WO2010147064A1

WO2010147064A1 - Magnetic stimulator

Info

Publication number: WO2010147064A1
Application number: PCT/JP2010/059969
Authority: WO
Inventors: 敦西川; 洋一齋藤; 大志福島; 正樹関野; 国克内田
Original assignee: 国立大学法人大阪大学
Priority date: 2009-06-15
Filing date: 2010-06-11
Publication date: 2010-12-23
Also published as: EP2444119B1; EP2444119A1; JP5622153B2; US10286222B2; JPWO2010147064A1; EP2444119A4; US20120157752A1

Abstract

Disclosed is a magnetic stimulator for magnetically stimulating a specific area of a subject. The magnetic stimulator is provided with a magnetic field generating means that has at least a stimulation coil and generates a dynamic magnetic field for applying magnetic stimuli, a coil holder that is manipulated to change the position of the magnetic field generating means relative to the specific area of the subject, a plurality of magnetic field sensors that detect the magnetic field generated by the magnetic field generating means, and a user interface that, either before or during the magnetic stimuli, provides instructions, which indicates the displacement that needs to be performed by the coil holder; the instructions are generated on the basis of the results from the detection of the magnetic field generated by the magnetic field generating means with the magnetic field sensors.

Description

Magnetic stimulator

The present invention relates to a magnetic stimulation apparatus for applying magnetic stimulation to a specific part of a subject (for example, a patient or a test examinee).

In recent years, interest in transcranial magnetic stimulation therapy has increased as a treatment method for many patients with neurological diseases for which drug treatment is not always effective. In this transcranial magnetic stimulation therapy, treatment and / or alleviation of symptoms can be achieved by applying magnetic stimulation to a specific part of the brain (for example, nerve in the brain) by a magnetic field generation source placed on the scalp surface of the patient. Unlike conventional electrical stimulation methods that use indwelling electrodes that require craniotomy and have a very strong resistance to the patient, this is a relatively new treatment that is non-invasive and requires less burden on the patient. It is expected to spread.

As a specific method of such transcranial magnetic stimulation therapy, an electric current is passed through a coil placed on the surface of the patient's scalp to locally generate a small pulsed magnetic field, and the principle of electromagnetic induction is used in the skull. A method is known in which an eddy current is generated to stimulate a nerve in the brain immediately below the coil (see, for example, Patent Document 1).
In this patent document 1, it is confirmed that refractory neuropathic pain is effectively reduced by transcranial magnetic stimulation treatment performed by such a method, and more accurate local stimulation realizes a higher pain reduction effect. Has been. However, it has also been clarified that the optimal stimulation site varies slightly depending on the individual patient.

Therefore, in order to obtain a higher effect by transcranial magnetic stimulation therapy, how to identify the optimal stimulation site of the patient's head for each individual patient, that is, the exact 3 of the stimulation coil for the patient's head. It is important how to perform dimension positioning. It is also known that even when the position of the stimulation coil is the same, a difference occurs in the effect obtained depending on its orientation (posture).
As for the positioning of the stimulation coil, a configuration in which the stimulation coil is positioned with respect to the patient's head using, for example, an optical tracking system using infrared rays is known (see, for example, Patent Documents 2 and 3). Some are already commercially available and clinically applied. Further, Patent Document 4 discloses an apparatus for positioning a stimulation coil with respect to a patient's head using an articulated robot.

Moreover, in the said patent document 1, when the above-mentioned transcranial magnetic stimulation treatment is performed, it will be clarified that the pain alleviation effect lasts for several hours but does not last for several days or more. ing. Therefore, it is considered desirable from the viewpoint of pain reduction to perform the above-mentioned therapy continuously every day if possible without leaving much time interval. In order to be able to perform such continuous treatment without forcing various physical and temporal burdens on the patient, it is possible to perform treatment at home or in a nearby clinic. Ideally.

International Publication No. 2007/123147 JP 2003-180649 A JP 2004-000636 A JP 2006-320425 A

However, the transcranial magnetic stimulation devices including the above conventional coil positioning devices and systems are assumed to be used in relatively large hospitals and research institutions for examination and research by skilled specialists. Therefore, handling and operation are complicated, skill is required to use, and it is quite large and expensive. For this reason, it is generally difficult for a patient, their family, or a nearby doctor who is not necessarily a specialist to operate and treat the patient. In addition, the cost is low at the patient's home or a relatively small clinic or clinic. Not only is the burden excessive, but it is generally difficult to secure an installation space.

Therefore, patients who undergo transcranial magnetic stimulation treatment still have to go to a large hospital where a large-scale magnetic stimulation device is installed for each treatment and there are skilled specialists, etc. In order to receive treatment repeatedly, the actual situation is that a large burden is imposed on various aspects.

In view of such circumstances, the present invention is easy to handle and operate so that a patient can perform transcranial magnetic stimulation therapy on a daily basis at home or in a nearby doctor's office. The basic object is to provide a magnetic stimulation device that is smaller and less expensive.

For this reason, the magnetic stimulation apparatus according to the present invention is a magnetic stimulation apparatus for applying a magnetic stimulation to a specific part of a subject, and a) a dynamic magnetic field generating means for generating a dynamic magnetic field for applying the magnetic stimulation. A magnetic field generating means including at least: b) an operating means operable to displace a relative position of the magnetic field generating means with respect to a specific part of the subject; and c) detecting a magnetic field generated by the magnetic field generating means. A plurality of magnetic field detecting means; d) prior to or during the magnetic stimulation, based on the detection result of the magnetic field generated from the magnetic field generating means by the magnetic field detecting means, the operation means should be used. And a notifying means for notifying teaching information for teaching a displacement operation.

A magnetic stimulation apparatus according to another invention of the present application is a magnetic stimulation apparatus for applying a magnetic stimulation to a specific part of a subject, and a) a dynamic magnetic field for generating a dynamic magnetic field for applying the magnetic stimulation. A magnetic field generating means including at least a generating means; b) a holding means for holding the magnetic field generating means in the vicinity of a specific part of the subject; and c) a plurality of magnetic field detecting means for detecting the magnetic field generated by the magnetic field generating means. D) Prior to or during the magnetic stimulation, the subject performs magnetic stimulation on the specific part based on the result of detection of the magnetic field generated from the magnetic field generation means by the magnetic field detection means. And a notifying means for notifying teaching information for teaching a body movement to be performed.

In the above case, it is preferable that a fixing means for fixing the magnetic field detecting means at a predetermined relative position with respect to the specific part of the subject is provided.
The magnetic field generation means is preferably attached to the operation means.

In the above case, the notifying means calculates the position of the magnetic field generating means as the position of the magnetic field source obtained by the inverse analysis method using each information on the magnetic field strength and direction detected by the plurality of magnetic field detecting means. Thus, the teaching information can be generated and notified. When applying the inverse analysis method, a random walk search method can be used.

Alternatively, information on the position of the magnetic field generation means and information on the magnetic field intensity and direction detected by the magnetic field detection means are recorded in pairs at a plurality of the positions in advance as the magnetic field generated at the position. The recording means further includes a recording means for comparing the information on the magnetic field intensity and direction detected by the magnetic field detection means prior to or during the magnetic stimulation with the recording information of the recording means. Based on this, the position of the magnetic field generating means may be calculated, and the teaching information may be generated and notified.

Alternatively, each of the magnetic field strengths and directions detected by each of the magnetic field detecting means in a state where the magnetic field generating means is located in a position for applying a magnetic stimulus to the specific part of the subject or in an allowable vicinity thereof. And further comprising target information recording means for previously recording a plurality of information, wherein the informing means includes information relating to the magnetic field strength and direction detected by the magnetic field detection means prior to or during the magnetic stimulation, and the target information. The teaching information may be generated and notified based on the comparison reference result with the recording information of the recording means.

The recording means can record the position information of the magnetic field generating means as position information in a plurality of different coordinate systems, and the position information in the plurality of different coordinate systems is mutually aligned to You may make it provide the coordinate conversion means for enabling contrast reference.

The informing means corresponds to at least the position of the dynamic magnetic field generating means in the specific part when performing magnetic stimulation for treatment on the specific part of the subject (more preferably, in the position and posture). Teaching for teaching the operation of the displacement to be performed using the operation means based on the deviation between the corresponding reference magnetic field data and the magnetic field data detected by the magnetic field detection means during the operation of the magnetic field generation means It can be configured to broadcast information.

In this case, the reference magnetic field data and at least the position reference data corresponding thereto are combined into a reference data set, and the notification means corresponds to at least the position of the dynamic magnetic field generation means. The magnetic field data based on the result detected by the magnetic field data is compared with the magnetic field data of a plurality of data sets including the reference data set, and the operation means is based on the position data of the data set that minimizes the difference between the two magnetic field data. The teaching information for teaching the operation of the displacement to be performed can be notified using.

In this case, a plurality of data sets other than the reference data set may be acquired separately from the acquisition of the reference data set.

In the above case, the magnetic field generating means can be configured to generate a dynamic magnetic field and a static magnetic field.
Alternatively, the magnetic field generating means can be configured to generate only a dynamic magnetic field.
Further, the magnetic field detection means can be configured to detect a dynamic magnetic field and a static magnetic field generated by the magnetic field generation means. In this case, the magnetic field detection means can be configured to detect the static magnetic field by the magnetic field generation means in a state where the generation of the dynamic magnetic field by the magnetic field generation means is stopped.
Alternatively, the magnetic field detection means can be configured to detect only the dynamic magnetic field generated by the magnetic field generation means.

A magnetic stimulation device according to still another embodiment of the present invention is a magnetic stimulation device for applying a magnetic stimulation to a specific part of a subject, and a) a motion that generates a dynamic magnetic field for applying the magnetic stimulation. A magnetic field generating means; b) an operating means to which at least a dynamic magnetic field generating means including this dynamic magnetic field generating means is attached and operated so as to be displaceable with respect to the specific part of the subject; c) generated by the magnetic field generating means A plurality of magnetic field detection means for detecting a magnetic field; d) a fixing means for fixing a position of the magnetic field detection means with respect to the specific part of the subject; and e) generation of the magnetic field based on detection signals of the plurality of magnetic field detection means. A magnetic field analysis means for inversely analyzing the magnetic field generated by the means to obtain three-dimensional data of the magnetic field generation means; f) a data storage means for storing required three-dimensional reference data of the magnetic field generation means; g) Comparison means for comparing the three-dimensional data obtained by the magnetic field analysis means with the three-dimensional reference data; h) based on the comparison result of the comparison means, the three-dimensional data from the three-dimensional reference data Informing means for informing teaching information for teaching the operation of displacement of the operating means in accordance with the deviation is provided.
The “required three-dimensional reference data of the magnetic field generating means” includes three-dimensional data of the magnetic field generating means corresponding to the optimum position and posture to which the magnetic stimulation should be applied in the specific part of the subject.

In the above case, the operation means may have only a dynamic magnetic field generation means as a magnetic field generation means, and the magnetic field detection means may be configured to detect a dynamic magnetic field generated by the dynamic magnetic field generation means.
Alternatively, the operating means includes a dynamic magnetic field generating means and a static magnetic field generating means as the magnetic field generating means, and the magnetic field detecting means stops the generation of the dynamic magnetic field by the dynamic magnetic field generating means. It can also be configured to detect a static magnetic field generated by the magnetic field generating means.

The three-dimensional reference data can be obtained by using a dedicated positioning device outside the magnetic stimulation device. As this “dedicated positioning device outside the apparatus”, for example, an optical tracking system or the like can be cited, which is required only when collecting reference data.
Alternatively, the three-dimensional reference data can be obtained using the magnetic field analysis means of the magnetic stimulation apparatus.

When performing reverse analysis of the magnetic field generated by the magnetic field generation means, the magnetic field analysis means preferably performs reverse analysis of the magnetic field by applying a random walk search method.

In the above invention, the notifying means for notifying the teaching information is preferably one notifying at least one of visual information and auditory information.
In particular, when the notification means notifies the teaching information by auditory information, at least one of the volume, the scale, and the timbre depends on the displacement amount to be performed by the operation means or the movement amount to be moved by the subject. It is preferable to notify one by changing one.
Alternatively, in the case where the notification means is a notification for notifying the teaching information by visual information, the teaching color is changed according to the displacement amount to be performed by the operation means or the movement amount to be moved by the subject. It is preferable to notify.

Moreover, as the fixing means, glasses, earphones, headphones, a headband, or the like can be suitably used.

The magnetic stimulation device as described above can be used as a device for applying magnetic stimulation to at least a specific part of the brain of a subject for transcranial magnetic stimulation treatment.

According to the present invention, in order to teach the displacement operation to be performed using the operation means based on the result of detection of the magnetic field generated from the magnetic field generation means by the magnetic field detection means prior to or during the magnetic stimulation. The teaching information is notified by the notification means.
Accordingly, the user (user) of the apparatus only operates based on the teaching information notified by the notification means, and does not require special skill as in the prior art, and the displacement to be performed using the operation means. Can be operated. That is, it can be operated and used relatively easily even by a patient, their family, or a nearby doctor who is not necessarily specialized. Moreover, since it is not necessary to use a large-scale and expensive apparatus as in the past, the cost burden is small, and it is easy to secure an installation space at the patient's individual home or a relatively small clinic or clinic. .
As described above, according to the present invention, it is possible to provide a magnetic stimulation device that is simple and easy to handle and operate, and that is smaller and less expensive. This makes it possible for a patient to perform daily routines at home or in a nearby clinic. Thus, it is possible to perform transcranial magnetic stimulation therapy continuously and repeatedly.

In the magnetic stimulation apparatus according to another aspect of the present invention, the magnetic field generation is performed by inversely analyzing the magnetic field generated by the magnetic field generation means attached to the operation means based on the detection signals of the plurality of magnetic field detection means. Magnetic field analysis means for obtaining three-dimensional data of the means is provided, and the three-dimensional data obtained by the magnetic field analysis means is compared with the three-dimensional reference data by the comparison means, and based on the comparison result, the three-dimensional data Teaching information for teaching the operation of displacement of the operating means according to the deviation from the three-dimensional reference data is notified by the notifying means.
Therefore, the user (user) of the apparatus can perform special skill as in the conventional case only by displacing the operation unit so that the deviation becomes zero (zero) based on the teaching information notified by the notification unit. The three-dimensional position and posture of the magnetic field generation means corresponding to the required three-dimensional reference data of the magnetic field generation means (that is, corresponding to the optimum position and posture to which the magnetic stimulation is to be applied) can be detected fairly easily. be able to. That is, it can be operated and used relatively easily even by a patient, their family, or a nearby doctor who is not necessarily specialized. Further, since it is not necessary to use a large-scale and expensive apparatus as in the prior art to detect the three-dimensional position and orientation of such magnetic field generating means, the cost burden can be reduced, and the patient's individual home or relatively small It is easy to secure installation space even in large clinics and clinics. In this way, it is possible to provide a magnetic stimulation device that is easy to handle and operate, and that is smaller and less expensive. This makes it possible for patients to repeat daily routines at home or in their nearby clinics. Transcranial magnetic stimulation therapy can be performed.

It is explanatory drawing which shows roughly the whole structure of the transcranial magnetic stimulation apparatus which concerns on one Embodiment of this invention. It is a perspective view which shows an example of the coil for stimulation used in the said embodiment, and a coil holder. It is a block block diagram which shows roughly the structure of the magnetic field analysis unit used by the said embodiment. It is a flowchart for demonstrating the operating method of the said transcranial magnetic stimulation apparatus at the time of the initial medical treatment performed in the hospital where a specialist doctor exists. It is a flowchart for demonstrating the operating method of the said transcranial magnetic stimulation apparatus at the time of home treatment. It is explanatory drawing which shows typically the magnetic field order analysis of a circular coil. It is explanatory drawing which shows typically the magnetic field reverse analysis of a circular coil. It is explanatory drawing which shows the vector which concerns on the exact solution in the magnetic field order analysis of a circular coil. It is explanatory drawing which shows typically the division | segmentation model for obtaining the approximate solution in the magnetic field order analysis of a circular coil. It is explanatory drawing which shows the 8-shaped spiral coil used in the said embodiment. It is explanatory drawing which shows typically the trust region method in a coil position search. It is a flowchart for demonstrating the random walk search method in a coil position search. It is explanatory drawing which shows typically the coordinate display at the time of using a permanent magnet. It is explanatory drawing which shows typically the coordinate display at the time of using the said 8-shaped spiral coil. It is a diagram for demonstrating the coordinate transformation at the time of using the said 8 shaped spiral coil. It is explanatory drawing which shows an example of the positional relationship of the generation | occurrence | production range of the assumption coil initial position and sensor position in the single coil reverse analysis simulation 1. FIG. It is a graph which shows the ratio of the trial frequency in which the difference | error with the true value was less than 10 [cm] in the said simulation 1. It is a graph which shows the ratio of the trial frequency in which the error with the true value was less than 5 [cm] in the said simulation 1. It is explanatory drawing which shows an example of the coil position search route in the said simulation 1. It is a graph which shows the minimum value and minimum value of the function f at the time of passing through the search route of FIG. It is a graph which shows the ratio of the frequency | count of trial in which the difference | error with a true value was settled within 1 [mm] in the simulation 2. FIG. 10 is a flowchart for explaining an execution procedure of a simulation 3; It is a graph which shows the correlation with the function f of a least square method, and the position error of a coil. It is a graph which expands and shows a part of Drawing 23A. It is a graph which shows the correlation with the said function f and the roll angle | corner error of a coil. It is a graph which expands and shows a part of Drawing 24A. It is a graph which shows the correlation with the said function f and the pitch angle error of a coil. It is a graph which expands and shows a part of Drawing 25A. It is a graph which shows the correlation with the said function f and the yaw angle error of a coil. It is a graph which expands and shows a part of Drawing 26A. The initial value of the search starting f and (f ₀₎ is a graph showing the correlation between the coil position specific permission. It is a graph which shows the ratio of convergence about the pattern whose error with an optimal attitude | position is zero. It is a graph which shows the ratio of convergence about the pattern whose error with an optimal attitude | position is in the range of +/- 10 [deg]. It is a graph which shows the ratio of convergence about the pattern whose error with respect to an optimal attitude | position is random within the range of +/- 20 [deg]. It is a graph which shows the ratio of convergence about the pattern whose error with respect to an optimal attitude | position is random within the range of +/- 30 [deg]. It is a perspective view which shows an example of the coil for stimulation used in experiment of embodiment (2nd Embodiment) which concerns on the method using a data set, and a coil holder. It is a perspective view which shows an example of the magnetic sensor fixing tool used in the experiment of the said 2nd Embodiment. It is a block block diagram which shows roughly the structure of the data set analysis unit used by the said 2nd Embodiment. It is explanatory drawing which shows the image display example of the screen of the display apparatus used by the said 2nd Embodiment. It is a flowchart for demonstrating the operating method of the magnetic stimulation apparatus performed in the hospital in the said 2nd Embodiment. It is a flowchart for demonstrating the operation method of the magnetic stimulation apparatus at the time of home treatment in the said 2nd Embodiment.

Hereinafter, an embodiment of the present invention will be described with reference to the accompanying drawings.
FIG. 1 is an explanatory diagram schematically showing the overall configuration of the transcranial magnetic stimulation apparatus according to the present embodiment. In this figure, a transcranial magnetic stimulation apparatus (hereinafter, abbreviated as “magnetic stimulation apparatus” or simply “apparatus” as appropriate), the entirety of which is indicated by numeral 10, is fixedly seated on a therapeutic chair 2. Treatment and / or symptom alleviation are performed by applying magnetic stimulation to nerves in the brain by the stimulation coil 11 placed on the scalp surface of the patient M (subject) who has been performed.

The stimulation coil 11 generates a dynamic magnetic field for applying a magnetic stimulation to at least a specific part of the brain of the patient M, and can be operated to be displaceable with respect to the head surface of the patient M. 12 is attached.
In FIG. 1, after the coil holder 12 is gripped and the stimulation coil 11 is displaced along the patient's scalp and the coil 11 is positioned, the coil 11 is not moved carelessly. More preferably, a state in which the coil holder 12 is fixed to the holder fixture 3 is shown.

As the stimulation coil 11, various types of known magnetic coils can be used. FIG. 2 is a perspective view showing an example of a stimulation coil and a coil holder that can be used as a dynamic magnetic field generating means in the transcranial magnetic stimulation therapy of the present embodiment. The stimulation coil 11 shown in FIG. 2 is a so-called eight-shaped spiral coil in which two spiral coils are arranged in the shape of a number “8” on the same plane. The induced current density becomes maximum just below the corresponding point. This type of magnetic coil 11 is somewhat difficult to fix including the identification of its posture, but is suitable for providing localized stimulation. The stimulation coil 11 is preferably integrally molded with the coil holder 12 when the synthetic resin coil holder 12 is molded.

The stimulation coil 11 is electrically connected to a magnetic stimulation control device 16 via a cable 15. The magnetic stimulation control device 16 controls the supply of current pulses to the stimulation coil 11, and various types of conventionally known types can be used. The operator performs an on / off operation of the magnetic stimulation control device 16. The operator can also set the current pulse intensity and pulse waveform that determine the intensity and cycle of the magnetic stimulation.

In the present embodiment, one magnetic field sensor 13 as a magnetic field detecting means capable of detecting the magnetic field generated by the stimulation coil 11 is attached to each of the left and right frame portions of the glasses 14 worn by the patient M.
Examples of the sensor 13 include an inductive sensor such as a so-called search coil, a Hall sensor using a Hall effect, an MR sensor using a magnetoresistance effect, an MI sensor using a magnetic impedance (Magneto-impedance), Furthermore, various types of known magnetic field sensors (magnetic sensors) such as a fluxgate type sensor can be used. Many mass-produced products with a size of several millimeters (mm) square and a weight of several grams (g) can be expected to be obtained at a price of about several hundred yen per piece. It can be said that a sufficiently small size, light weight and low price can be achieved for use in transcranial magnetic stimulation therapy.

The eyeglasses 14 serve as fixing means for fixing the positions of a plurality (for example, two in this embodiment) of the magnetic field sensors 13 with respect to the patient's head. The reproducibility is required for the fixed position of the magnetic field sensor 13 with respect to the head of the patient M, and it is necessary to always fix the magnetic field sensor 13 to the same position with respect to the patient M. As a means for fixing the magnetic field sensor 13 to the patient's head relatively naturally without causing discomfort or discomfort to the patient M, the opportunity to be worn and used on a daily basis is familiar. An orthosis (body wear) is desirable. In this regard, the glasses 14 are suitable. It should be noted that although there is a case where the position shift slightly occurs in general glasses, so-called protective (safety) glasses and sports glasses are designed so that such position shift hardly occurs. It is particularly suitable as the 13 mounting fixtures.

The transcranial magnetic stimulation apparatus 10 according to the present embodiment performs reverse analysis on the detected magnetic field (magnetic field strength and direction) based on the detection signals of the plurality of magnetic field sensors 13 and obtains a magnetic field source obtained by this inverse analysis method. And calculating and acquiring three-dimensional data of the stimulation coil 11 that has generated the magnetic field as at least a position (preferably a position and a posture) of the first and second reference data (three-dimensional reference data) described later. In contrast, a magnetic field analysis unit 20 capable of detecting a deviation (deviation) from the three-dimensional reference data is provided.

Further, the magnetic field analysis unit 20 can display the deviation detected by the unit 20 to a user (for example, an operator) by, for example, visual information, for example, a display device 28 including a liquid crystal type display panel. Is attached.
The display device 28 performs a reverse magnetic field analysis to grasp the current position (preferably the current position and posture) of the stimulation coil 11, and then notifies the user of the current position (and posture) of the coil 11 for stimulation. It serves as an interface for guiding the coil 11 to the optimal position (that is, the position corresponding to the optimal stimulation site) and posture. In this case, the “user” is, for example, a patient, his / her family, a doctor such as a doctor's office, a medical worker, or the like.

The magnetic field analysis unit 20 includes, for example, a so-called personal computer including a CPU (Central Processing Unit) as a main part. As shown in the block configuration diagram of FIG. 3, the signal analysis unit 22, the storage unit 23, A comparison unit 24 and a user information output unit 25 are provided.
The signal analysis unit 22 is preferably based on detection signals input from the plurality of magnetic field sensors 13 (sensor 1, sensor 2,..., Sensor N) input as wireless signals (see FIG. 3: arrow Y1). The magnetic field generated by the stimulation coil 11 is inversely analyzed to obtain three-dimensional data of the stimulation coil 11, that is, three-dimensional data regarding the position and orientation of the stimulation coil 11.

In this specification, “the posture of the stimulation coil” means the direction and angle of the stimulation coil 11, and “the direction of the stimulation coil” means the direction of the coil 11 on the scalp surface of the patient M. That is, “the angle of the stimulation coil” means an angle formed by the normal of the scalp surface of the patient M and the magnetic field direction of the coil 11.

In addition, the storage unit 23 has three-dimensional data of the stimulation coil 11 corresponding to the optimal position and posture to which magnetic stimulation should be applied on the head of the patient M (that is, three-dimensional data on the position and posture of the stimulation coil 11). ) As reference data (see FIG. 3: arrow Y2), and is composed of a readable memory device.
This three-dimensional reference data indicates that when a magnetic stimulation is applied to a specific part of the brain of the patient M using the stimulation coil 11, the optimal coil position (so-called sweet spot) at which the neuropathic pain of the patient M is most reduced. And the posture, which can be determined using a dedicated positioning device outside the transcranial magnetic stimulation device 10 when performing medical treatment in a hospital such as during initial medical treatment.

Examples of the “dedicated positioning device outside the device” include a conventionally known optical tracking device and an optical tracking system including a medical image (both not shown). When the three-dimensional reference data is collected, Only needed.
Alternatively, instead of using such a “dedicated positioning device outside the apparatus”, the magnetic field analysis unit 20 of the transcranial magnetic stimulation apparatus 10 is used, specifically, the signal analysis unit 22 of the magnetic field analysis unit 20. The three-dimensional reference data can be determined by using the function.

The comparison unit 24 compares the three-dimensional data obtained by the signal analysis unit 22 with the three-dimensional reference data stored in the storage unit 23 (see FIG. 3: arrows Y3 and Y4). A deviation (deviation) of the three-dimensional data obtained by the analysis unit 22 from the three-dimensional reference data can be detected.
Then, the deviation data detected by the comparison result in the comparison unit 24 is output as a signal to the user interface unit 28 (in the present embodiment, the above-described display device) via the user information output unit 25 (FIG. 3: See arrows Y5 and Y6). Based on the output signal from the user information output unit 25, the user interface unit 28 teaches information for teaching the displacement operation to be performed using the operation means (coil holder 12) (in the case of the display device described above). For example, a display signal such as a video signal is generated and notified to the user.

The operator (user) of the device 10 visually recognizes the display device 28 (see FIG. 3: arrow Y7), so that the deviation displayed on the screen of the display device 28 becomes zero (zero) as much as possible. 12 is displaced along the scalp of patient M. Then, the displacement operation of the stimulation coil 11 is stopped at the position and posture of the stimulation coil 11 at which the deviation displayed on the screen of the display device 28 is zero (zero) or as close to zero as possible. Hold. At this time, as shown in FIG. 1, it is convenient to fix the coil holder 12 using the holder fixture 3.

The operation method of the transcranial magnetic stimulation apparatus 10 configured as described above is divided into an initial medical treatment performed by a specialized doctor at the hospital and a home treatment performed by the patient M or his family at home, as shown in FIG. This will be described with reference to the flowchart of FIG.
First, an operation method of the apparatus 10 at the time of initial medical care performed in a relatively large hospital where a specialized doctor is present will be described based on a flowchart of FIG.

This hospital has a conventionally known optical tracking device and medical image display as a dedicated device outside the device for positioning the coil 11 for magnetic stimulation used for transcranial magnetic stimulation therapy to the optimum position (and posture). It is assumed that an optical tracking system (hereinafter referred to as “reference measurement system” as appropriate) including an apparatus (both not shown) is provided. By using the optical tracking system, the coil is checked while confirming the target based on a medical image (for example, MRI (magnetic resonance imaging) image) of the brain of the patient M displayed on the medical image display device. For example, nerves in the brain can be magnetically stimulated while monitoring the position and direction / angle with respect to the brain surface in real time.

When the operation of the apparatus 10 is started during the initial medical examination, first, the patient M wears the sensor unit 13 in Step # 11. That is, the patient M wears spectacles 14 to which a plurality (two in this embodiment) of magnetic field sensors 13 are attached. Accordingly, detection signals are input from the plurality of magnetic field sensors 13 (sensor 1, sensor 2,..., Sensor N) to the signal analysis unit 22 (see FIG. 3: arrow Y1). Next, in step # 12, the doctor looks at the optical tracking device and the medical image (both not shown), and refers to the muscle reaction in the area where the patient M feels pain, while referring to the stimulation coil 11. The optimum position and posture are searched, and it is continuously determined whether or not the stimulation coil 11 has reached the optimum position and posture (step # 13).

When the stimulation coil 11 reaches the optimum position and posture and the determination result in step # 13 is YES, a trigger input from the doctor (switch: ON / see FIG. 3: arrow Y8) is input at that time. Based on the detected signal of the magnetic field sensor 13, the position and orientation of the stimulation coil 11 are calculated as three-dimensional data by the signal analysis unit 22 (step # 14). That is, three-dimensional data in a sensor system (hereinafter referred to as a sensor measurement system as appropriate) corresponding to the optimum position and posture of the stimulation coil 11 is obtained. As shown by an arrow Y2 in FIG. 3, the three-dimensional data obtained in step # 14 is stored in the storage unit 23 as reference data (step # 15). As described above, when the three-dimensional reference data is set in the initial medical care performed at the hospital, the comparison unit 24, the user information output unit 25, and the user interface unit 28 do not operate.

Next, the operation method of the apparatus 10 at the time of home treatment performed by the patient M or his / her family at home will be described based on the flowchart of FIG. During home treatment, the magnetic field analysis unit 20 of the transcranial magnetic stimulation apparatus 10 stores three-dimensional reference data in the storage unit 23 in advance by the initial treatment in the hospital described with reference to the flowchart of FIG. It shall be. Therefore, at the time of this home treatment, both the trigger input from the doctor indicated by the arrow Y8 in FIG. 3 and the storage of the three-dimensional reference data associated therewith in the storage unit 23 (see FIG. 3: arrow Y2) are performed together. There is nothing.

When the operation of the apparatus 10 is started during treatment at home, first, the patient M wears the sensor unit 13 in step # 21. That is, the spectacles 14 to which a plurality (two in this embodiment) of magnetic field sensors 13 are attached are worn. Accordingly, detection signals are input from the plurality of magnetic field sensors 13 (sensor 1, sensor 2,..., Sensor N) to the signal analysis unit 22 (see FIG. 3: arrow Y1). Next, in step # 22, the patient M or his / her family or the like grips the coil holder 12 so that the stimulation coil 11 is positioned in the optimal position and posture as much as possible along the scalp surface of the patient M. Operate displacement.

Corresponding to the displacement of the stimulation coil 11, the position and orientation of the stimulation coil 11 are calculated as three-dimensional data by the signal analysis unit 22 in step # 23. The three-dimensional data calculated by the signal analysis unit 22 and the three-dimensional reference data stored in the storage unit 23 are compared by the comparison unit 24 (see FIG. 3: arrows Y3 and Y4). A deviation (deviation) of the three-dimensional data obtained by the signal analysis unit 22 from the three-dimensional reference data is detected. Then, the deviation data detected by the comparison result in the comparison unit 24 is output as a signal to the user interface unit 28 (the display device in the present embodiment) via the user information output unit 25 (see FIG. 3: See arrows Y5 and Y6).

While viewing the image displayed on the display device 28 (see FIG. 3: arrow Y7), the patient M or his / her family or the like holds the coil holder 12 so that the deviation displayed on the screen is zero (zero) as much as possible. A displacement operation is performed along the scalp of patient M. That is, the patient M or his / her family is navigated to the optimum position and posture of the stimulation coil 11 by the user interface unit 28 (display device) (step # 24).
At this time, according to the magnitude of the deviation of the three-dimensional data obtained by the signal analysis unit 22 from the three-dimensional reference data (the amount of displacement to be performed by the coil holder 12), that is, the stimulation coil 11 is at the optimum position. As the color approaches, the color of the image projected on the display device 28 is changed, for example, from blue to yellow, and further to red as the deviation becomes smaller. 11 to the optimum position and posture becomes easier, and the convenience can be further enhanced.

Then, in step # 25, it is determined whether or not the stimulation coil 11 has reached the optimum position and posture. While this determination result is NO, each step after step # 22 is repeatedly executed. Thereafter, when the deviation displayed on the screen of the display device 28 is zero (or as close to zero) as possible and the determination result in the step # 25 is YES, the position and posture of the stimulation coil 11 at that time Then, the displacement operation of the stimulation coil 11 is stopped and the state is maintained. At this time, as described above, it is convenient to fix the coil holder 12 using the holder fixture 3 (see FIG. 1).

As described above, according to the present embodiment, the stimulation coil 11 (magnetic field generation means) attached to the coil holder 12 (operation means) based on the detection signals of the plurality of magnetic field sensors 13 (magnetic field detection means). The signal analysis unit 22 (magnetic field analysis means) of the magnetic field analysis unit 20 that obtains the three-dimensional data of the stimulation coil 11 by inversely analyzing the magnetic field generated by the The three-dimensional data is compared with the three-dimensional reference data stored in the storage unit 23 by the comparison unit 24 (comparison means). Based on the comparison result, the deviation of the three-dimensional data from the three-dimensional reference data is Notification is made by the interface unit 28 (notification means).

Therefore, a user (user) of the magnetic stimulation apparatus 10 can move the coil holder 12 so that the deviation notified by the user interface 28 becomes zero (zero). The three-dimensional position of the stimulation coil 11 corresponding to the required three-dimensional reference data of the stimulation coil 11 (that is, corresponding to the optimum position and posture to which the magnetic stimulation should be applied) and The posture can be detected. In other words, the patient M or his / her family or a nearby doctor who is not necessarily specialized can operate and use relatively easily. Further, since it is not necessary to use a large-scale and expensive device as in the prior art to detect the three-dimensional position and posture of the stimulation coil 11, the cost burden can be reduced, and the patient's home or relatively It is easy to secure installation space even in small clinics and clinics. That is, it is possible to provide a magnetic stimulation apparatus 10 that is simple to handle and operate, and that is smaller and less expensive, so that the patient M can continue to repeat daily at home or in a nearby clinic. Transcranial magnetic stimulation therapy can be performed.

In the description of the operation method described above, in the initial medical treatment process in the hospital shown in the flowchart of FIG. 4, the step in which the doctor searches for the optimal position and posture of the stimulation coil 11 (FIG. 4: step # 12) It is assumed that the measurement is performed using a known optical tracking system (reference measurement system). Then, three-dimensional data in the sensor measurement system corresponding to the optimum position and posture of the stimulation coil 11 specified (FIG. 4: step # 13) using this reference measurement system is calculated (FIG. 4: step # 14). This is stored in the storage unit 23 as three-dimensional reference data (FIG. 4: step # 15). However, the sensor system (sensor measurement system) and the optical tracking system (reference measurement) attached to the patient M are used. Since the registration with the system) can be easily performed as will be described later, in a simpler version, the “three-dimensional sensor measurement system corresponding to the optimum position and orientation of the stimulation coil 11” in step # 14 of FIG. “Calculation of data” is unnecessary, and the three-dimensional data of the optimal position and orientation of the stimulation coil 11 specified by the reference measurement system is stored in the storage unit 23 as it is. It may be.

Even in this configuration, the three-dimensional data in the reference measurement system for the optimal position and orientation of the stimulation coil 11 specified by the doctor can be re-baked into the sensor measurement system by simple coordinate conversion. It can be used without particular trouble as three-dimensional reference data during treatment.
However, if errors that may occur due to coordinate transformation must be taken into account, as shown in steps # 14 and # 15 of FIG. 4, three-dimensional data in the sensor measurement system is also calculated. However, it is necessary to store this in the storage unit 23 as three-dimensional reference data.

The “registration” has various known methods, and an example thereof will be described. For example, when coordinate conversion from the coordinate system B to the coordinate system A is performed, a coordinate conversion matrix can be derived by the following basic procedure, and coordinate conversion can be performed using this.
(1) First, four feature points that are not on the same plane that can be stably measured are determined.
(2) Next, the position coordinates of the four feature points as seen in the coordinate system A:
(X ₁ , y ₁ , z ₁ ), (x ₂ , y ₂ , z ₂ ), (x ₃ , y ₃ , z ₃ ), (x ₄ , y ₄ , z ₄ )
To get.
(3) Further, the position coordinates of the same four feature points as seen in the coordinate system B:
(X ₁ , Y ₁ , Z ₁ ), (X ₂ , Y ₂ , Z ₂ ), (X ₃ , Y ₃ , Z ₃ ), (X ₄ , Y ₄ , Z ₄ )
To get.

(4) The coordinate transformation matrix T is calculated by Equation 1.

(5) By using this coordinate transformation matrix T, as shown in Equation 2, the position coordinates (x, Y, Z) of an arbitrary feature point acquired in the coordinate system B as seen in the coordinate system A (x , Y, z).

Here, the reason why the coordinate transformation matrix T can be calculated by Equation 1 will be described. Since the four feature points seen in the coordinate system A and the four feature points seen in the coordinate system B are each physically the same point, Equation 3 is established.

Since four feature points that are not on the same plane are selected, there is an inverse matrix of the matrix shown in Equation 4, and by multiplying both sides of Equation 3 from the right, Equation 1 is obtained. It is done.

Registration of the above-described sensor system (sensor measurement system: coordinate system A) and optical tracking system (reference measurement system: coordinate system B) performed using such a method will be described.
(1) First, four feature points that are not on the same plane that can be stably measured are determined on a magnetic field sensor mounting device (in this case, for example, glasses 14).
(2) Next, the position coordinates of the four feature points as seen in the coordinate system A:
(X ₁ , y ₁ , z ₁ ), (x ₂ , y ₂ , z ₂ ), (x ₃ , y ₃ , z ₃ ), (x ₄ , y ₄ , z ₄ )
To get. In this case, since the coordinate system A is a coordinate system fixed to the magnetic field sensor mounting device (glasses 14), the position coordinates of the four feature points can be obtained from the design values of the magnetic field sensor fixture (glasses 14). it can.
(3) Furthermore, the position coordinates of the same four feature points as seen in the coordinate system B using a hospital three-dimensional position measuring device (optical tracking system):
(X ₁ , Y ₁ , Z ₁ ), (X ₂ , Y ₂ , Z ₂ ), (X ₃ , Y ₃ , Z ₃ ), (X ₄ , Y ₄ , Z ₄ )
To get.
(4) The coordinate transformation matrix T is calculated by the above equation 1.
(5) By using this coordinate transformation matrix T, as shown in Equation 2, the position coordinates (X of the optimum stimulus position) of an arbitrary feature point acquired by the coordinate system B (optical tracking system) , Y, Z) can be converted into position coordinates (x, y, z) in the coordinate system A.

In the description of the operation method described above, in the initial medical treatment process in the hospital shown in the flowchart of FIG. 4, the data stored in the storage unit 23 is the optimal position and posture of the stimulation coil 11 specified by the doctor. In addition to the three-dimensional reference data in the corresponding sensor measurement system, for example, the coil 11 is placed at a head position where the patient M is supposed to place the stimulation coil 11 during home treatment. The sensor detection value and the three-dimensional data of the position and orientation of the coil 11 (data obtained by converting the data acquired by the reference measurement system into the sensor measurement system) are also stored in the storage unit 23 at the same time. It is also possible to increase the speed and / or accuracy of the analysis program.
In this case, with regard to the head position where the patient M is supposed to place the stimulation coil 11 during home treatment, the position can be easily identified by dividing the head into a mesh and displaying the coordinates. That's fine.

Further, in the description of the operation method described above, in the initial medical care process in the hospital shown in the flowchart of FIG. 4, the doctor searches for the optimal position and posture of the stimulation coil 11 (FIG. 4: step # 12). Is assumed to be performed using a known optical tracking system (reference measurement system), but without using such an optical tracking system, the magnetic field analysis unit 20 of the apparatus 10 is used for stimulation. It is also possible to search for the optimum position and posture of the coil 11.

As is well known, an optical marker is generally used in a known optical tracking system. However, when the magnetic field analysis unit 20 is applied, a permanent magnet may be used instead of the optical marker. Using this permanent magnet, like a normal calibration, instead of the reflective marker at the position of the nose or ear, apply the magnetic marker, and reversely analyze with the sensor system to obtain the magnetic marker position coordinates, Registration of the relationship between sensor coordinates and medical image coordinates becomes possible. This registration can be similarly performed by applying the above-described method.

In the above embodiment, the eyeglasses 14 are used as an attachment device to which the magnetic field sensor 13 is attached. However, other body accessories such as earphones, headphones, and headbands can be used instead.
There are many types of headbands that use various materials to be worn along the shape of the user's forehead and its vicinity. In recent years, earphones and headphones have also been used by users' ears. Are commercially available and can be suitably used as a mounting device for the magnetic field sensor 13.

In the above embodiment, the user interface unit 28 serving as a notification unit that notifies the deviation of the three-dimensional data obtained by the magnetic field analysis unit 20 from the three-dimensional reference data is notified by visual information. For example, the display device 28 having a liquid crystal type display panel has been used, but instead of this, or in addition to this, the deviation can be notified by auditory information from a speaker or the like. In this case, according to the magnitude of the deviation (the amount of displacement that the coil holder 12 should perform), that is, as the stimulation coil 11 approaches the optimum position, at least one of volume, scale, and timbre changes. By making it comprise, the induction | guidance | derivation to the optimal position and attitude | position of the coil 11 for a stimulus becomes easier, and the convenience can be improved more.

Furthermore, in the above embodiment, the so-called 8-shaped spiral coil suitable for providing the limited magnetic stimulation is used as the stimulation coil 11, but for example, the spot to which the magnetic stimulation is applied is relatively large ( In such a case, a simple circular coil that can be easily handled, including specifying the coil posture, and that can easily perform magnetic field analysis can be preferably used.
Note that the stimulation coil 11 is not limited to that disclosed in the present specification, and various types of stimulation coils may be used depending on the purpose of magnetic stimulation, the strength of stimulation required, and various other factors. Can be used.

In the above embodiment, the three-dimensional reference data corresponding to the optimal position and posture of the stimulation coil 11 obtained in advance at the initial medical examination and the magnetic field inverse analysis of the stimulation coil 11 obtained at home treatment are used. The coil 11 is guided to the optimum position and posture by comparing with the three-dimensional data and navigating by visual information or auditory information according to the latter deviation with respect to the former. It is also possible to directly compare the magnetic field information and guide the coil 11 to the optimum position and orientation instead of obtaining and comparing the three-dimensional data from the two.

For example, as described above, the optimum position and posture of the stimulation coil 11 can be searched using the magnetic field analysis unit 20 of the apparatus 10 without using an optical tracking system. The magnetic field data to be the reference obtained from 13 and the magnetic field data obtained from the magnetic field sensor 13 at home treatment are directly compared without converting them into three-dimensional data, respectively, and the latter (at home treatment) magnetic field data. Is guided as close as possible to the magnetic field data of the former (at the time of initial medical treatment), and even if the actual three-dimensional position and orientation of the coil 11 are not known, the magnetic field information is used as it is. Can be guided to an optimal position and posture.

Furthermore, in the above embodiment, the magnetic field sensor 13 as the magnetic field detection means detects the magnetic field generated by the stimulation coil 11 as the dynamic magnetic field generation means attached to the coil holder 12 as the operation means. However, in addition to the stimulating coil 11, a permanent magnet as a static magnetic field generating means is provided in the coil holder 12 (operation means), and this permanent magnet can be used exclusively for positioning. In this configuration, the magnetic field sensor 13 detects not only a dynamic magnetic field but also a static magnetic field.

In this case, by detecting the static magnetic field generated by the permanent magnet while the generation of the dynamic magnetic field by the stimulation coil 11 is stopped, the interference of the dynamic magnetic field is avoided and the static magnetic field generated by the permanent magnet is detected. A positioning signal by a magnetic field can be accurately detected. In particular, in transcranial magnetic stimulation therapy, since the generation and stop of a dynamic magnetic field are repeated intermittently during treatment, it is conceivable to turn on the magnetic field sensor in synchronization with the timing of stopping the dynamic magnetic field. However, it is possible to detect a static magnetic field (permanent magnet) and perform magnetic field analysis without stopping the generation of a dynamic magnetic field during treatment etc. by providing a plurality of magnetic field sensors with different sensitivities. is there.
Furthermore, during stimulation (during treatment), the dynamic magnetic field is generated with instantaneous pulses (for example, 10 times per second), so it is conceivable to switch ON / OFF of the magnetic field sensor in synchronization with the pulse timing. By turning off the magnetic field sensor during pulse generation, interference of the dynamic magnetic field can be avoided. In this case, sensors having different sensitivities may not be provided.

In addition, even when both coil positioning and magnetic stimulation treatment are performed using only a dynamic magnetic field without using a static magnetic field, a relatively weak dynamic magnetic field is generated prior to magnetic stimulation treatment to position the coil. Thereafter, the coil position can be finely adjusted while performing a magnetic stimulation treatment by generating a relatively strong dynamic magnetic field. In this case, the magnetic field sensor 13 needs to detect two types of dynamic magnetic fields, a relatively weak dynamic magnetic field and a relatively strong dynamic magnetic field.
From the above, the following three types with different magnetic field strengths are considered as types of magnetic fields to be detected by the magnetic field sensor.
(A) A relatively weak dynamic magnetic field (generated prior to magnetic stimulation treatment)
(B) A relatively strong dynamic magnetic field (generated mainly for magnetic stimulation treatment)
(C) Static magnetic field by a permanent magnet (introduced exclusively for positioning)

In the same magnetic field sensor, it is practically impossible to detect magnetic fields having different magnetic field strengths. In such a case, for example, it is convenient if a plurality of magnetic field sensor groups having different sensitivities as described below are provided.
(1) For example, as described above, a magnetic field sensor group for detecting a permanent magnet marker for alignment with medical image information during initial treatment in a hospital: In this magnetic field sensor group, Detect magnetic field.
(2) Magnetic field sensor group for positioning coils prior to magnetic stimulation treatment when used at home: This magnetic field sensor group detects a relatively weak dynamic magnetic field (a) or a static magnetic field (c). To do.
(3) Magnetic field sensor group for finely adjusting the coil position during magnetic stimulation treatment when used at home: In this magnetic field sensor group, (b) and (c) Detect the synthesized magnetic field.

By providing a plurality of magnetic field sensor groups with different sensitivities as described above, coil positioning can be performed by detecting magnetic fields with different magnetic field strengths without changing the principle of coil position detection or its specific algorithm. It can be performed.

Next, a method for specifying the three-dimensional position of the stimulation coil 11 performed by the signal analysis unit 22 of the magnetic field analysis unit 20 will be described.
In order to specify the three-dimensional position of the stimulation coil 11, “magnetic field reverse analysis” is required, and this magnetic field reverse analysis requires magnetic field forward analysis. As is well known, the “magnetic field forward analysis” is to analyze a magnetic field signal at an arbitrary place where the position of the magnetic field generation source is known (see FIG. 6). Is for analyzing the position of the magnetic field generation source with known magnetic field signals at a plurality of locations (see FIG. 7).

<Field analysis method>
First, the magnetic field sequence analysis method will be described taking a simple circular (annular) coil as an example.
As shown in FIG. 8, it is assumed that a circular coil having a radius a is in a plane perpendicular to the z-axis with the origin as the center, and a current I is supplied to the coil. At this time, the magnetic field vector generated by the circular coil is as follows depending on whether the solution is an exact solution or an approximate solution. Here, μo is the magnetic permeability of vacuum, and the unit is MKSA unit system.

[Exact solution]
Assuming that the magnetic field vector at an arbitrary point (r, θ, φ) shown in FIG. 8 is B = (B _r , B _θ , B _φ ), each component by the exact solution is expressed by Equation 5.
Here, _Aφ is a vector potential and is expressed by Equation 6, K (k) and E (k) are the first and second type complete elliptic integrals, respectively, and k is expressed by Equation 7.

[In case of approximate solution]
In obtaining an approximate solution, as shown in FIG. 9, a model in which a circular coil is divided into N pieces in the circumferential direction is assumed, and each divided element is approximated by a straight line (line segment). In this divided model, the magnetic field generated at the position r by the nth element at the position r ′ is given by Bio-Savart's law expressed by Equation 8. Here, “Δs”, “r ′”, and “t (r ′)” in Expression 8 are given by the following equations, respectively.
Δs = 2πa / N
T (r ′) = (− sin θn, cos θn, 0)
R '= (a.cos .theta.n, a.sin .theta.n, 0)
However, θn = 2πn / N

The magnetic field generated by the circular coil as a whole is the sum of the magnetic field vectors generated by each element, and is expressed by the following equation (9).

The magnetic field distribution approximated based on the above is expressed by the following

equations

10 and 11.

As described above, the coil used for the magnetic stimulation according to the present embodiment is not a simple circular shape (annular shape), and as schematically shown in FIG. It is a coil formed side by side in the shape of "8". There is a shape in which two spiral coils are bent so as to form a mountain shape with a predetermined angle between them.
Therefore, in the present embodiment, the magnetic field distribution is forwardly analyzed using an approximate solution method that can be applied to the difference in coil shape. Since the law of superposition holds for the magnetic field, when the magnetic fields generated from the left and right spiral coils are superimposed and expressed, the following equation 12 is obtained. In Equation 12, the first term represents the contribution of the left spiral, and the second term represents the contribution of the right spiral. In Equation 12, K represents the number of turns of the coil, a _o represents the outer radius dimension, a _i represents the inner radius dimension, and h represents the distance between the centers of the two coils.

<Inverse magnetic field analysis method>
Next, a magnetic field inverse analysis method will be described. In the present embodiment, in the magnetic field inverse analysis that specifies the three-dimensional position of the coil from the magnetic field signal, the least square method is used in addition to the magnetic field forward analysis expressed by Equation 12.

[Least square method]
In this case, n nonlinear equations with n variables are assumed as follows.
F ₁ (x ₁ , x ₂ ,..., X _n ) = 0
F ₂ (x ₁ , x ₂ ,..., X _n ) = 0
......
・ F _n (x ₁ , x ₂ ,..., X _n ) = 0
Here, a set of n variables [x ₁ , x ₂ ,..., X _n ] is represented by x, and a set of n functions [f ₁ (x), f ₂ (x) _,. (X)] can be expressed as f (x), and the above-mentioned n nonlinear equations can be expressed as f (x) = 0.

Since this equation f (x) = 0 is a nonlinear equation, there is almost no so-called exact solution. Therefore, some approximation is required to solve this equation. As one of the approximations, instead of solving the above equation, use the size of f (x), that is, F (x) = ∥f (x) ∥ ² , and minimize x ^* One possible method is to find the value. If F (x ^* ) is sufficiently close to zero (zero), this x ^* may be considered as a solution. This method is a so-called least square method.

In this embodiment, assuming that the magnetic field read by the α-th sensor is Bα, and the coil is present at a certain position (coordinates are known), and the magnetic field at the sensor position obtained by forward analysis is B, The equation can be expressed by the following equation (13).

B is updated so that the value of f is minimized, that is, the assumed coil position is updated. This updating method will be described below.
The hypothetical coil position that gives B when f is minimum is approximated to the actual coil position, and the current coordinates (three-dimensional data) of the stimulation coil can be obtained. Based on the three-dimensional data of the current position, the coil position is searched and updated so that the deviation (deviation) from the optimum position (three-dimensional reference data) specified at the time of initial medical care becomes zero.

[Coil position update method 1]
As one of the coil position update methods (that is, search methods), it is conceivable to use a so-called dog leg type trust region method. This method is a method developed to compensate for this drawback in view of the problem that the Newton method does not converge due to the initial value, and that the global convergence is guaranteed and the convergence is good. Has been. However, the algorithm is more complicated and the analysis time is longer than the Newton method.
As shown in FIG. 11, the trust area reduction method is a method constructed by adding a trust region to the Newton method and the steepest descent method.

B _k is the Hessian matrix ∇ ² f (x _k ) or an approximate matrix thereof, the confidence radius is Δ _k, and the second-order model function q _k (s _k ) of the following equation in which B _k is a positive definite value (restriction condition: | S _k | ≦ Δ _k ) and Δ _k > 0.
q _k (s _k ) = f (x _k ) + ∇f (x _k ) ^T s _k + 1/2 · s _k ^T B _k s _k
At this time, a point obtained as the Newton method is set as x _N = x _k −Bk ⁻¹ ∇f (x _k ), and 2 when moving from the point x _k along the steepest descent direction −∇f (x _k ) The minimum point of the next model q _k (s _k ) is set as x _cp . This x _cp is called a Cauchy point. If x _N is inside the trust region, then let x _k + s _k = x _N , otherwise, a piecewise linear point connecting x _k , x _cp , x _N and from x _k pick point distance is delta _k as _x _k + _s k (see FIG. 11). That is, when the confidence radius is sufficiently large, the Newton method is adopted, and when the confidence radius is small, a method considering the steepest descent direction is adopted.

Furthermore, the decrease amount of the model function value Δq _k = q _k (s _k ) −q _k (0) = ∇f (x _k ) ^T s _k + ½ · s _k ^T B _k s _k
And a decrease amount of the objective function value Δf _k = f (x _k + s _k ) −f (x _k )
And the approximate solution is updated as appropriate based on the magnitude of these reductions (the update condition is arbitrary). Further, the size of the trust region is appropriately changed according to the situation at that time.

[Coil position update method 2]
It is conceivable to use a random walk (RW) search method as another coil position update method (that is, a search method).
In this RW method, for example, as shown in the flowchart of FIG. 12, the vicinity of the assumed coil position is randomly selected, and the value of f is calculated by the least square method when it is assumed that the coil has moved to that position. If the value of is improved, the coil position is updated. By repeating this, it is a relatively simple method of gradually converging to the correct coil position.

Each parameter in the RW method shown in the flowchart of FIG. 12 represents the following matters.
N _th : upper limit threshold for the number of radius searches (n) r _th : lower limit threshold for the search radius (r) α: radius update parameter (where α <1)

In the flowchart of FIG. 12, the radius search count n = 0 in the initial state. When the search is started from this initial state, first, in step # 51, the value of f at the assumed coil current position is calculated. The value of f at this time (that is, the initial value of f) is set to f ₀ (f = f ₀ ). Next, in step # 52, the value of f at a random position separated from the “current position” in step # 51 by a radius r is calculated. Then, at step # 53, the calculated value f determines whether less than the initial value _{_{f 0 (f <f 0)}} . If the determination result in step # 53 is YES, the value of f has improved, so the position is updated (step # 54). That is, the search is successful.

On the other hand, of the case where the determination result is NO in step # 53, in step # 55, whether the radius search number n has not reached its upper threshold n _th (n <n _th) is determined, not reach If not (step # 55: YES), in step # 58, the counter value of the radius search number n is incremented (that is, the counter value n is increased by 1), and the process returns to step # 52 and each of the subsequent steps. Repeat steps. Further, if the determination in step # 55 is NO, since the radius search times n has reached the upper limit threshold value n _th, in step # 56, the counter is reset radius search times n (n = 0) At the same time, the search radius r is set to αr (r = αr), and at step # 57, it is determined whether or not the search radius r has reached the lower limit threshold _rth (r <r _th ). If the determination result in step # 57 is NO, the process returns to step # 52 and the subsequent steps are repeated. On the other hand, if the decision result in the step # 57 is YES, the search radius r has reached the lower limit threshold value _rth , and therefore the search has failed.

<Coordinate transformation>
As shown in FIG. 13, in the case of a permanent magnet, the magnetic field does not change even if the magnet rotates around the axis connecting the north and south poles. However, as shown in FIG. In the case of the 8-shaped coil used in the form, the magnetic field detected by the magnetic field sensor changes even if it rotates about any of the U, V, and W axes. In other words, in the case of a permanent magnet, it is possible to perform an inverse analysis if two coordinates of the N pole and the S pole are known. In the case of an 8-shaped coil, however, it can be further analyzed in the plane where the coil is placed. If you don't know the coordinates, you can't do reverse analysis. Further, in the case of an 8-shaped coil, forward analysis is possible only in the coil coordinate system (UVW coordinate system), so the sensor position must be converted to a new coil coordinate system each time the assumed coil position is updated. . It can be said that this is a difficult point of the coil magnetic field inverse analysis.

Next, coordinate conversion of a three-dimensional coordinate system required for coil magnetic field inverse analysis will be described.
As shown in FIG. 15, it is assumed that two three-dimensional orthogonal coordinate systems Σ _a and Σ _b are set. Here, a point representing the point P in the coordinate system Σ _a is ^a p, ^{a point} p representing the coordinate system Σ _b is ^b p, and a vector representing the position vector of the origin of the coordinate system Σ _{b in} the coordinate system Σ _a is ^a Let q _b . At this time, the relationship between ^a p and ^b p is expressed by the following equation.
^a p = ^a R _b ^b p + ^a q _b
Here, ^a R _b is a 3 × 3 matrix representing the attitude of the coordinate system Σ _b with respect to the coordinate system Σ _a and is called a rotation matrix. Since the column vectors is representing a unit vector in each axis of the coordinate system sigma _b, the following equation holds.
^{_{^{_{b R a = (a R b}}}} ) T = a R b -1

Therefore, the relationship between ^a p and ^b p is also expressed by the following equation (14).
The rotation matrix ^a R _b around each of the X, Y, and Z axes is expressed by Equation 15 when θ rotates around the X axis, and by Equation 16 when φ rotates around the Y axis. In the case of ψ rotation about the Z axis, these are respectively expressed by Expression 17.

Various simulations were performed using the inverse analysis method described above.
Next, this inverse analysis simulation will be described.
<Simulation 1>
First, as a first stage of the simulation, a so-called single coil having one turn is assumed for the sake of simplicity, and the coil posture is fixed (that is, the coil coordinate system is simply translated from the absolute coordinate system). In this case, it was verified whether or not the exact coordinates of the coil could be back-analyzed. More specifically, the extent to which the convergence changes due to the difference in the initial position of the assumed coil (that is, the initial position dependence of the convergence) is verified, and the difference in the convergence due to the difference in the number of magnetic field sensors is also verified. did. The coil has a total of 6 degrees of freedom with 3 degrees of freedom according to position and orientation. Therefore, if the number of magnetic field sensors is theoretically two, the inverse analysis is possible. Here, a simulation was performed for each of the cases where the number of sensors was 2, 3, and 4.

In this simulation, the actual center position of the coil is x = 0.90 to 1.10 [m], y = 0.90 to 1.10 [m], z = 1.10 to 1.20 [m]. Randomly determined from the rectangular parallelepiped represented by The sensor coordinates were set as shown in

Equations

18, 19, and 20 for the cases where the number of sensors was 2, 3, and 4, respectively.

The generation pattern of the assumed initial coil position was set as shown in Table 1.

FIG. 16 shows the relationship between the generation range of the assumed coil initial position, the actual range of the center position of the coil, and the sensor position, taking the pattern S3 in Table 1 as an example.
For updating the position of the hypothetical coil, only the trust region method was used, and in each pattern, the hypothetical coil initial position was changed 100 times, and the convergence was verified. In each trial, the upper limit threshold of the number of calculations by the trust region method was set to 30 times.

[Results and discussion of simulation 1]
Table 2 shows an error (that is, a minimum error) at the time of the trial that approaches the true value among the 100 trials. FIG. 17 shows the ratio of the number of trials in which the error from the true value is within 10 [cm] among 100 trials by pattern, and FIG. 18 shows that the error from the true value is also 5 [cm]. The ratio of the number of trials falling within the range is represented by pattern.

From the data shown in Table 2, FIG. 17 and FIG. 18, when the assumed coil initial position was started from a place closer to the true value (patterns S2, S4, S6), it was started from a place far from the true value. It was found that the convergence was clearly better than the time (patterns S1, S3, S5). That is, it can be said that the convergence in performing the magnetic field inverse analysis has an initial position dependency.
However, when the assumed initial coil positions are the same and the number of sensors is different (patterns S1, S3, and S5, and patterns S2, S4, and S6), in the coil magnetic field inverse analysis according to this simulation, the convergence by the number of sensors No effect was observed.

Further, as described above, in each trial, the upper limit threshold of the number of calculations by the trust region method was set to 30. However, even if the number of calculations was further increased, there was no significant change in the error from the true value. . This is considered to be because in each trial, the value of f obtained by the least square method at each final point takes a minimum value. That is, it is considered that the function f has a plurality of minimum values at a place several centimeters from the true value. In the trust region method, if the number of calculations is increased, the solution always converges to the minimum value without diverging. However, if the function f has a plurality of minimum values near the true value, the magnitude of the error from the true value is small. Regardless of the case, there arises a problem that it converges to any minimum value (not the minimum value).

<Simulation 2>
[Verification of existence of local minimum]
In the present embodiment, the magnetic field generated by the coil is weak when it is about 4 to 5 cm or more away from the coil, and rapidly increases when it is closer to it. For this reason, as shown in FIG. 19, once the search path for the assumed coil position by the trust region method is closer to the magnetic field sensor than the actual coil position, it is obtained by the magnetic field and magnetic field sensor obtained by the forward analysis. As shown in FIG. 20, the function f has a local minimum that is not a minimum (global minimum), as shown in FIG. Note that it is known that the trust region method tends to approach a true value through a route as shown in FIG.

FIG. 20 is a plot of the value of f in 1 [mm] increments in each axial direction when the three points represented by Equation 21 are linearly moved in order. Since the actual coil is between the second point and the third point, the value of f is zero at the corresponding position. The sensor arrangement was the same as in the case of the number of sensors = 4 in the simulation 1 described above.

As described above, as an improvement measure for preventing the minimum value (not the minimum value) from being taken, it is conceivable to dispose the magnetic field sensor at a sufficiently distant position. The mounting position is restricted, and it is actually difficult to dispose at a sufficiently distant position. In addition, if the magnetic field sensor is separated too much, it is likely that the magnetic field sensor is easily affected by disturbances, which is not preferable.

[Introduction of random walk search method]
In transcranial magnetic stimulation therapy, if the position of the coil deviates from the optimum position by, for example, about 5 [mm] or more, the therapeutic effect may be diminished. In such a case, the error from the true value should be targeted within at least 5 [mm]. However, as described above, only the trust region method converges to a minimum value, and it is generally considered difficult to satisfy the requirement. The above simulation results also show this.
The so-called “Simulated Annealing” method is known as an algorithm for converging to the minimum value instead of the local minimum value. However, in this method, there is generally a practical difficulty in setting parameters. It has been known.

Therefore, it was considered to introduce a random walk search method (RW method) that is relatively easy to implement. That is, the algorithm uses the RW method from the position where the update by the trust region method ends (that is, the position within a few [cm] from the true value).
In the simulation 2, the convergence was verified with each pattern of Table 1 in the simulation 1 using this algorithm. Each parameter in the RW method was set as follows.
_-Upper limit threshold of radius search number n: n _th = 10
· Search radius r of the lower _threshold: r th = 0.1 [mm]
・ Radius update parameter: α = 0.9
-Initial value of search radius r: r ₀ = 10 [mm]

[Results and discussion of simulation 2]
FIG. 21 shows the ratio of the number of trials in which the error from the true value within 100 trials falls within 1 [mm] for each pattern. As can be seen from FIG. 21, as in the case of simulation 1, when the assumed initial coil position is started from a place close to the true value (patterns S2, S4, S6), it starts from a place far from the true value. The pattern S2, S4, and S6, where the initial position of the assumed coil is started from a location close to the true value, is clearly better than the case (patterns S1, S3, and S5). The convergence rate is over 70%.
Therefore, if the calculation is performed two or three times, it can be said that the position of the single coil can be estimated without considering the posture of the coil by combining the trust region method and the RW method.

In this treatment, the patient usually knows the optimum position of the patient to some extent, and it is considered that the initial position where the stimulation coil is placed is not greatly mistaken. Therefore, the magnetic field inverse analysis basically has no problem only with the RW method, and the application of the trust region method may be limited to the case where the initial position where the patient puts the coil is largely mistaken. Since the trust region method includes second-order partial differential calculation, which increases the burden on calculation man-hours, it can be said that it is preferable not to use at all from such a viewpoint.

<Simulation 3>
In this simulation 3, the stimulation coil was changed from the above-described single coil to the 8-shaped spiral coil according to the present embodiment, and a simulation was performed in consideration of the coil posture.
As described above, when the patient moves the stimulation coil, it is considered that the patient does not deviate greatly from the optimum position instructed by the specialist in the initial medical treatment (that is, has a certain degree of reproducibility). It is possible to set the initial position generation space in the vicinity of the optimum position. Therefore, in this simulation 3, the trust region method is not used, and only the RW method is applied. The XYZ axes were set as the absolute coordinate system, and the UVW axis was set as the coil coordinate system. The sensor arrangement was the same as in the case of the number of sensors = 4 in the simulation 1 described above.

[Execution procedure]
Next, the execution procedure of the simulation 3 will be described with reference to the flowchart of FIG.
When the simulation starts, first, in step # 71, an actual coil position is set. In this simulation 3, it was assumed that the patient placed the coil at a position slightly deviated from the optimum position instructed in the initial medical examination and at a position slightly deviated from the optimum position. Specifically, the actual coil position is set at a point translated from the optimal position by the initial deviation s ₀ [cm], and further within the range of β [deg] for each of the U axis, V axis, and W axis. The posture rotated at random was assumed to be the actual coil posture.

In step # 72, the search is started. In this initial search state, the number of searches i = 0, and the assumed coil initial position change count j = 0. When the search is started from this initial state, first, the assumed coil position is set to the optimum position (step # 73), and the coil attitude is rotated around each of the U, V, and W axes (coordinate conversion). And the posture is such that the value of f is minimized (step # 74). Next, the RW method is applied (step # 75), and it is determined whether or not the calculated value of f at the assumed coil current position is smaller than the upper limit threshold f _th (f <f _th ) (step #). 76). If the determination result in step # 76 is YES, the search is successful.

On the other hand, if the determination result in step # 76 is NO, it is determined in step # 77 whether the number of searches i has not reached the upper limit threshold i _th (i <i _th ). If (step # 77: YES), in step # 80, the search number i counter is incremented (that is, the counter value i is increased by 1), and the process returns to step # 74 and the subsequent steps are repeated. . If the determination result in step # 77 is NO, the number of searches i has reached the upper limit threshold i _th , and therefore the assumed coil initial position is changed in step # 78.

In this step # 78, the assumed initial coil position is set to a position shifted from the optimum position by, for example, s ₁ = 5.0γ [cm] (a random constant in the range of γ: 0 <γ <1), The counter for the number of searches i is reset (i = 0), and the counter for the number j of changes in the assumed coil initial position is incremented (that is, the counter value j is increased by 1). Then, further, in step # 79, it is determined whether non reached its upper threshold _{j th} number of changes j assumptions coil initial position (j _{<j th).} If the decision result in the step # 79 is YES, the process returns to the step # 74 and the subsequent steps are repeated. On the other hand, if the decision result in the step # 79 is NO, since the number of changes j assumptions coil initial position it has reached the upper threshold j _th, a search fails, abandon the search.

[Evaluation of optimal position (determination of f _th )]
In the above simulation, since the actual coil position is grasped, whether or not it has converged can be determined by comparing the coil position / posture at the end of the search with the actual coil position / posture. It was. However, in the actual treatment, since the actual coil position (position where the patient hits the coil) is not known, it is necessary to determine whether or not the convergence has been achieved only by the f-value by the least square method. Therefore, it is necessary to verify how small the value of f should satisfy the required specifications, that is, how much the threshold value f _th in the algorithm described in FIG. 22 should be set.

Here, the required specification is preferably calculated based on clinical research, and in this embodiment, the coil center position and posture errors are set to be within the following ranges, for example.
-Coil center position error: within 5 [mm]-Coil posture error: within 5 [deg] for each coil axis

FIG. 23A is a correlation diagram showing the correlation between the value of f by the least square method and the error from the optimum position, and FIG. 23B is an enlarged view of a part of FIG. 23A. 24A and 24B are correlation diagrams showing the correlation between the value f and the roll angle error, FIGS. 25A and 25B are correlation diagrams showing the correlation between the value f and the pitch angle error, and FIGS. 26A and 26B. FIG. 4 is a correlation diagram showing a correlation between the value of f and a yaw angle error. Here, the roll angle refers to the rotation angle around the U axis shown in FIGS. 14 and 15, and the pitch angle and the yaw angle refer to rotation angles around the V axis and the W axis, respectively. And

As shown in FIGS. 23A and 23B to FIGS. 26A and 26B, for any error, as the value of f by the least square method increases, the error increases, and between the value of f and each error. Was found to have a positive correlation.
23A and 23B, it can be said that the condition that the error from the optimum position of the value of f by the least square method satisfies the required specification is approximately f <5.0 × 10 ⁻⁸ . Further, from FIGS. 24A and 24B to FIGS. 26A and 26B, it can be said that the condition that the error from the optimum posture of the value f satisfies the required specifications is approximately f <1.5 × 10 ⁻⁷ . Therefore, it can be seen that the coil position can be specified almost accurately by setting f _th = 5.0 × 10 ⁻⁸ .

[Reliability of specified coil position and orientation]
The smaller the f _th value is set, the higher the reliability of the position and orientation of the identified coil. On the other hand, the average time required for the reverse analysis is generally longer. It is also important from a practical point of view to shorten the time required for analysis processing while ensuring the necessary reliability according to the situation.
Since transcranial magnetic stimulation treatment is performed while pressing the stimulation coil against the patient's head, there is a tolerance for errors in the W-axis direction of the coil coordinate system (direction perpendicular to the coil plane: see FIG. 14). It is relatively large, and it is considered that the threshold can be raised a little with respect to the position error. Further, it can be seen from FIGS. 23A and 23B to FIGS. 26A and 26B that even if f _th ≧ 5.0 × 10 ⁻⁸ , the required specifications are satisfied with a considerable probability. Therefore, it is not a good idea to exclude the entire range of f _th ≧ 5.0 × 10 ⁻⁸ regardless of the situation.

For example, when the current position of the coil is far away from the optimum position, the reliability is slightly inferior to increasing the reliability of the coil position over time. In general, it is generally better to give priority to approaching the optimal position as soon as possible.
Therefore, for example, as shown in Table 3 below, the relationship between the value of f by the least square method and the reliability is defined, and the reliability and the reliability of the information along with the information on the position and orientation of the coil obtained by the inverse analysis are also defined. At the same time, it may be communicated to the patient. In transcranial magnetic stimulation treatment, even if a place slightly deviated from the optimum position is stimulated, in general, there is no problem in safety because the effect of treatment is diminished. For example, in the case of treatment based on information with low reliability, when the patient feels that the treatment effect is small, the patient selects information according to the reliability so that information with low reliability is not used. It is also possible to make it possible.

[Determination of the initial value (f ₀ ) of f]
In the course of proceeding with the above discussion on the magnetic field inverse analysis, if the value of f (initial value (f ₀ )) at the time of starting the search is larger than a certain level, the required specification is not satisfied with certainty (that is, The tendency to fail the search) was recognized. In such a case, if the initial value of f (f ₀ ) is larger than a certain threshold value, the search from that position is set not to be performed from the beginning (so-called filter is applied), so that the analysis time is reduced. Shortening becomes possible.

FIG. 27 shows the correlation between the initial value (f ₀ ) of f when starting the search and whether or not the coil position can be specified. The graph of FIG. 27 plots 50 pieces of f ₀ data when the required specification is not satisfied after searching by the RW method and 50 pieces of f ₀ data when the required specification is satisfied. From the graph of FIG. 27, by applying a filter that “search is not performed when f ₀ > 2.5 × 10 ⁻⁴ ”, an attempt to “success” the search is rarely omitted, but the search is “ It can be seen that about half of the trials that are “failed” are omitted, and an improvement in search efficiency can be achieved.

[Searchable range and convergence time]
Considering the above-mentioned f _th and f _0, to what extent the inverse analysis is possible, that is, to what extent “if the patient has placed the stimulation coil out of the optimal position, If it is “deviation”, it is verified whether the position of the coil can be specified.
In this verification, initial conditions were set as shown in Table 4. In each pattern shown in Table 4, the optimum position / posture was changed 100 times and a search trial was performed. To describe the case of Table 4 in a pattern T ₁₂ as an example, in the pattern T _12, place the coil at a position patient shifted 1 [cm] from the optimum position, and, ± a posture of the coil to each axis The search was performed by randomly shifting 100 times within a range of 20 [deg]. For other patterns, the method is the same except that the combination of the range of error s ₀ and error β is different. The verification was performed by setting the upper limit threshold (i _th ) of the number of executions of the RW method to 1000 times and the upper limit threshold (j _th ) of the number of times to change the initial search position to 10 times.

FIG. 28 to FIG. 31 are graphs showing the rate of convergence when a search trial is performed with the optimum position and orientation changed 100 times in each pattern.
More specifically, FIG. 28 shows a combination of three patterns (T ₁₀ , T ₃₀ , T ₅₀ ) in which the error β with respect to the optimum posture is zero (β = ± 0 [deg]), and FIG. 29 shows the optimum posture. FIG. 30 shows the optimum combination of three patterns (T ₁₁ , T ₃₁ , T ₅₁ ) that are random within the range of error β of ± 10 [deg] and have different errors s ₀ from the optimum position. The combinations of three patterns (T ₁₂ , T ₃₂ , T ₅₂ ), which are random within the range of error β from the attitude within ± 20 [deg] and have different errors s ₀ from the optimum position, respectively, the combination of 31 optimal attitude and error β is ± 30 [deg] error _{s 0} are different three patterns between the optimum position a random within the scope of _{_{_{(T 13, T 33, T}}} 53) Nitsu Te indicates the percentage of each convergence. In each figure, in the case of the reliability E, the ratio that the reliability is not equal to or higher than the reliability B is shown even if the search is performed by changing the initial search position 10 times.

From FIG. 28, it can be seen that if the error β with respect to the optimum posture is zero (β = ± 0 [deg]), the position of the coil can be specified with extremely high reliability. In addition, it can be seen from FIGS. 28 to 31 that the rate at which the position and orientation of the coil can be specified decreases as the distance from the optimum position increases. Regarding the initial posture, the pattern close to the optimal posture has a slightly higher convergence rate, but does not show any superiority. That is, it can be seen that the coil position error (s ₀ ) has a greater influence on the coil inverse analysis than the coil attitude error (β).

Since the ratio of the reliability T or higher (S + A + B) in the pattern T ₅₃ is about 50% (see FIG. 31), when the error s ₀ with respect to the optimum position exceeds 5 [cm], the ratio further decreases. Expected to be below The worse the convergence rate, the more time is required for inverse analysis and stress is given to the patient. Therefore, from the viewpoint of enabling the patient to smoothly guide the coil, the error s ₀ from the optimum position is preferably within 5 [cm].
It is also believed that the patient does not move the coil too rapidly. Therefore, once the position of the coil is specified, it can be assumed that the initial position of the hypothetical coil is given in the vicinity of the coil position. That is, it is considered that the inverse analysis can always be performed from the vicinity of the actual coil position, and the position of the coil can be grasped with high reliability.

Table 5 shows a convergence time when f _th <3.0 × 10 ⁻⁷ is satisfied and the position of the coil can be specified in each pattern of Table 4.

As shown in Table 5, a search time of 0.51 [sec] at the shortest and 8.90 [sec] at the longest is required. If this simulation does not converge after one search trial, the initial search position of the hypothetical coil is changed up to 10 times and the search is performed again. Therefore, if the initial position is changed a lot, more time is required for the search. It will take. When comparing the average search time, the patterns T ₁₀ , T ₃₀ , and T ₅₀ converge at 1 [sec] or less, and the other patterns converge at around 1.5 [sec]. . From this, it can be seen that although the convergence rate decreases as the initial position of the hypothetical coil moves away from the optimum position, when it converges in any pattern, it converges in approximately the same time.

As described above, according to the present embodiment, by using the above method, the magnetic field generated by the stimulation coil 11 is inversely analyzed based on the detection signals of the plurality of magnetic field sensors 13, and the stimulation is performed. The current position and orientation of the coil 11 can be specified. In particular, by applying the random walk search method, the function f in the least squares method can be converged to the minimum value instead of the minimum value, and the current position and posture of the stimulation coil 11 can be more reliably determined. Can be identified.
By enabling such a magnetic field inverse analysis, the magnetic stimulation device 10 can be reduced in size, cost, and operation, and the patient M can be routinely used at home or in a nearby clinic. Continuous and repeated transcranial magnetic stimulation therapy can be performed.

All of the above explanations correspond to the optimum position and posture to which the magnetic stimulation is to be applied by using an inverse analysis method using information on the magnetic field intensity and direction detected by the magnetic field detection means (magnetic field sensor 13). The stimulation coil is guided to the three-dimensional position and posture, but instead of this, or in combination with this, information on at least the position (more preferably, the position and posture) of the magnetic field generating means. By using a combination of data (data) and a combination of data (data) related to the magnetic field intensity and direction detected by the magnetic field detection means for the magnetic field generated at the position (in this specification, Such a combination of data is referred to as a “data set”), and a stimulation coil at a three-dimensional position (and posture) corresponding to the optimum stimulation position (and posture). It can be induced.

Hereinafter, an embodiment related to a technique using this data set (hereinafter referred to as a second embodiment as appropriate) will be described. In the following description, the same or similar configuration is basically the same as or similar to that in the embodiment using the above-described inverse analysis method (hereinafter referred to as the first embodiment as appropriate). The components having the same functions are denoted by the same reference numerals, and further description thereof is omitted.

FIG. 32 is a perspective view showing an example of a stimulation coil and a coil holder used in the experimental example of the present embodiment. As shown in this figure, the stimulation coil 11 used in the experimental example in the second embodiment is the same as that used in the first embodiment, and is a so-called “8-shaped” spiral coil. More preferably, when the synthetic resin coil holder 12 is molded, it is integrally molded with the coil holder 12. Needless to say, as in the case of the first embodiment, other various types of known magnetic coils can be used as the stimulation coil.

In the present embodiment, a permanent magnet serving as a static magnetic field generating means for position detection is provided on a predetermined portion of the coil holder 12, for example, on a surface portion corresponding to the outer side of both ends in the longitudinal direction of the stimulation coil 11. 41 and 41 are fixed, respectively.
Further, on the upper surface of the coil holder 12, a transparent base plate 42 made of, for example, resin is erected in a direction perpendicular to the longitudinal direction at a central portion in the longitudinal direction of the stimulation coil 11. It is fixed. The base plate 42 is preferably detachably fixed to the coil holder 12 using, for example, a screw member.

The base plate 42 is used as a detection target marker when the position of the coil holder 12 (that is, the position of the stimulation coil 11) is detected by a known optical tracking system (for example, POLALIS manufactured by NDI). A so-called polaris marker 43 is fixed. A plurality of the markers 43 are preferably provided. In this case, more preferably, the three spherical markers 43 are arranged on the base plate 42 so as to be positioned at the vertices of a triangle having a predetermined shape.

In the example of FIG. 32, static magnetic field generating means (permanent magnets 41 and 41) is used to detect the position of the coil holder 12 (that is, the position of the stimulation coil 11). By detecting the static magnetic field generated by the permanent magnet in a state where the generation of the dynamic magnetic field by the coil 11 is stopped, interference of the dynamic magnetic field is avoided, and a positioning signal by the static magnetic field generated by the permanent magnet is generated. It can be detected accurately. However, by providing a plurality of magnetic field sensors having different sensitivities, the position of the coil holder 12 is detected by detecting the static magnetic field (permanent magnet) without stopping the generation of the dynamic magnetic field during treatment or the like. It is also possible.
Alternatively, as described in the description of the first embodiment, position detection can be performed using dynamic magnetic field generation means (for example, the stimulation coil 11).

FIG. 33 is a perspective view showing an example of a magnetic field sensor fixture for fixing the magnetic field sensor at a predetermined relative position with respect to a specific part of a patient. The magnetic field sensor fixture 50 is configured by a U-shaped frame body in a plan view, and the frame body 50 is sized and shaped so that the inner edge of the frame body 50 is firmly locked to the patient's head. . In the experimental example of this embodiment, a mannequin head Hm was used instead of the patient's head.
A plurality of magnetic field sensors 51 are preferably fixed on the upper surface of the frame body 50. In this experimental example, a pair of front and rear magnetic field sensors 51 are respectively attached to the left and

right side portions

50a and 50b of the frame body 50, and a total of four magnetic field sensors 51 are used. As a result, the magnetic field can be detected (that is, the magnetic field strength and the direction of the magnetic field are detected) at four locations on the front, rear, left and right surrounding the head Hm. As the magnetic field sensor 51, a so-called triaxial sensor is preferably used. Instead of this, it goes without saying that various other types of known magnetic field sensors can be used as in the first embodiment.

In this experimental example, more preferably, a transparent plate-like base plate 52 made of, for example, resin is fixed to a predetermined portion of the front side portion 50c of the frame body 50 so as to hang down. 52, three spherical polaris markers 53 are attached as detection targets by the optical tracking system. The marker 53 is used only for error evaluation during the experiment.

The magnetic field sensor fixture 50 serves as a fixing means for fixing the position of a plurality (for example, four in this embodiment) of the magnetic field sensors 13 with respect to the patient's head. In the case of the first embodiment Similarly, the reproducibility is required for the fixed position of the magnetic field sensor 13 with respect to the patient's head, and it is necessary to always fix the magnetic field sensor 13 to the same position with respect to the patient. As a means of fixing the magnetic field sensor 13 to the patient's head relatively naturally without causing discomfort or discomfort to the patient, there are many opportunities to wear it on a daily basis. Familiar equipment (body wearing equipment) is desirable.
Therefore, in practical use, as in the first embodiment, it is preferable to use eyeglasses, particularly protective (safety) eyeglasses, sports eyeglasses, or body accessories such as earphones, headphones, and headbands.

In the example of FIG. 33, four magnetic field sensors 13 are used by being attached to the magnetic field sensor fixture 50, but other numbers of magnetic field sensors may be used. As is well known, it is generally preferable to increase the number of sensors in order to increase the measurement accuracy. However, the system becomes complicated and the cost increases accordingly. For this reason, it is desirable to reduce the number of sensors as much as possible, but even in that case, it is preferable to use at least two sensors in order to ensure a certain level of measurement accuracy. More preferably, the head is isotropically arranged.

If it demonstrates more concretely using the above coil holder 12 and the magnetic field sensor fixture 50, in this embodiment, as mentioned above, the permanent magnet 41 is attached to the coil holder 12 holding the coil 11 for stimulation, as mentioned above. At a hospital, the magnetic field sensor fixing tool 50 is attached to a patient, and magnetic field data (related to the strength and direction of the magnetic field) obtained by detecting the magnetic field generated by the permanent magnet 41 of the coil holder 12 with the magnetic field sensor 13 is obtained. Data) and three-dimensional position and orientation data of the coil holder 12 (that is, the stimulation coil 11) obtained by the optical tracking system, for example, are measured at the same time, and the combination of both data is made into one "data set" Record. The data set is collected and recorded for the optimal stimulation position specified by the doctor using a conventional method and a plurality of (many) positions around the optimal stimulation position.

An example of the data set collected in this way is shown in Table 6.
In this example, the number of magnetic sensors is four, and they are displayed using subscripts a, b, c, and d, respectively. The position / posture of the coil is represented by P for the center position of the coil and R for the posture. Data displayed using the subscripts 1 to N represent data corresponding to the data set numbers 1 to N, respectively. Since the magnetic sensor measures values in three directions of x, y, and z, the magnetic field data B _a to B _d are three-dimensional vectors, and the values in the respective directions are suffixed with x, y, and z. To display. Taking the case of data set number 1 as an example for a magnetic sensor displayed using the subscript a, the magnetic field data B _a1 is expressed by the following equation.
・ B _a1 = (B _a1x , B _a1y , B _a1z )
Similarly, the position data P and the attitude data R are also three-dimensional vectors. If the case of the data set number 1 is shown as an example, the position data P ₁ is expressed by the following equation.
・ P ₁ = (P _1x , P _1y , P _1z )
The attitude data R1 is represented by the following expression in the case of data set number 1 if the roll angle is represented by α, the pitch angle is represented by β, and the yaw angle is represented by γ.
R ₁ = (α ₁ , β ₁ , γ ₁ )

FIG. 34 is a block configuration diagram schematically showing the configuration of the data set analysis unit used in the second embodiment.
The data set analysis unit 120 includes, for example, a so-called personal computer having a CPU (Central Processing Unit) as a main part, and as shown in a block configuration diagram of FIG. 34, a signal analysis unit 122 and a recording unit 123. And a comparison unit 124 and a user information output unit 125.

The signal analysis unit 122 is preferably based on detection signals input from the plurality of magnetic field sensors 13 (sensor 1, sensor 2,..., Sensor N) input as radio signals (see FIG. 34: arrow Y1). Then, magnetic field data (data relating to the strength and direction of the magnetic field) generated by the permanent magnet 41 of the coil holder 12 is acquired and input to the recording unit 123 (see FIG. 34: arrow Y2). At the same time, for example, the doctor obtains the data of the three-dimensional position and posture of the coil holder 12 (that is, the stimulation coil 11) obtained by the optical tracking system and inputs the data to the recording unit 123 (FIG. 34: arrow Y8). reference).

The recording unit 123 records the combination of the magnetic field data obtained by the simultaneous measurement as described above and the data of the three-dimensional position and posture as one “data set”, and is specified by a doctor using a conventional method. The data set is collected and recorded with respect to the optimum stimulation position and posture and a plurality of (many) positions and postures around the optimum stimulation position and posture, and is configured by a readable memory device.

The comparison unit 124 compares the magnetic field data obtained by the signal analysis unit 122 with the data set recorded in the recording unit 123 during treatment (see FIG. 34: arrows Y3 and Y4). From the recorded data set, a data set having a magnetic field closest to the magnetic field data obtained by the signal analysis unit 122 (even if it does not completely match) is extracted. A deviation (deviation) from the three-dimensional position and posture data (three-dimensional reference data) corresponding to the optimum stimulation position and posture can be detected from the three-dimensional position and posture data of the data set. .

Then, the data of the three-dimensional position and orientation of the data set extracted based on the comparison result in the comparison unit 124 is output as a signal to the user interface unit 128 (display device in the present embodiment) via the user information output unit 125. (See FIG. 34: arrows Y5 and Y6). At this time, it is preferable that the three-dimensional data (three-dimensional reference data) of the optimal stimulus position and posture is input in advance to the user interface unit 128 (display device). The magnetic field data corresponding to the three-dimensional reference data is reference magnetic field data.
Based on the output signal from the user information output unit 125, the user interface unit 128 teaches information for teaching the displacement operation to be performed using the operation means (coil holder 12) (in the case of the display device described above). For example, a display signal such as a video signal is generated and notified to the user.

The operator (user) of the coil holder 12 visually recognizes the display device 128 (see FIG. 34: arrow Y7), and controls the coil so that the deviation displayed on the screen of the display device 128 becomes zero (zero) as much as possible. The holder 12 is displaced along the patient's scalp. Then, the displacement operation of the stimulation coil 11 is stopped at the position and posture of the stimulation coil 11 at which the deviation displayed on the screen of the display device 128 is zero (zero) or as close to zero as possible. Hold. At this time, as in the case of the first embodiment, it is convenient to fix the coil holder 12 using a holder fixture.

In the present embodiment, a graphic program interface Open GL is installed for displaying an image on the display device 128, and as shown in FIG. 35, a coil holder image 55 (FIG. 35) corresponding to the optimum stimulation position. : One-dot chain line) and the coil holder image 56 (FIG. 35: solid line) at the current position are displayed on the same screen. More preferably, the sensor fixture image 57 is also displayed in the same screen.
Therefore, the operator (user) of the coil holder 12 can see the coil holder image 56 (current position) of the solid line displayed on the screen while viewing the screen of the display device 128 while the coil holder image 55 (optimum of the dashed line) The coil holder 12 may be displaced along the patient's scalp so as to overlap the stimulation position as much as possible.

A method of operating the magnetic stimulation apparatus including the data set analysis unit 120 configured as described above will be described with reference to the flowcharts of FIGS.
First, in a hospital (see FIG. 36), in step # 101, the patient is caused to wear the magnetic field sensor fixture 50. At this time, calibration relating to the mounting position of the magnetic field sensor fixture 50 is performed (step # 102). Then, the doctor specifies the optimal stimulus position by a method using a conventional optical tracking system (step # 103). Thereafter, in step # 104, data sets are collected for the optimal stimulation position and a plurality of (many) positions in the vicinity thereof, and recorded in the recording unit 123 of the data set analysis unit 120.

Next, at home treatment, as in the case of the hospital, first, the patient is required to wear the magnetic field sensor fixture 50 (step # 121). At this time, the patient wears the magnetic field sensor fixture 50 in the same manner as in the hospital, but it is actually difficult to wear the magnetic field sensor fixture 50 without any positional deviation, so the magnetic field sensor fixture It is necessary to calibrate 50 mounting positions (step # 122). Then, by using the navigation function of the user interface unit 128 of the data set analysis unit 120, the coil holder 12 is moved and operated while viewing the display screen, and the stimulation coil is positioned as close as possible to the optimal stimulation position and posture. 11 is guided to perform treatment (step # 123). At this time, the base plate 42 (and the polaris marker 43) fixed to the coil holder 12 with a screw member or the like may be removed and the coil holder 12 may be moved. Further, the base plate 52 and the polaris marker 53 attached to the magnetic field sensor fixture 50 are not necessary.

The calibration relating to the mounting position of the magnetic field sensor fixture 50 performed in Step # 102 and Step # 122 can be performed as follows, for example. In this case, preferably, a certain mounting position within the desired range is determined as a reference mounting position so that the magnetic field sensor fixture 50 is mounted at a position within the desired range, and the patient is set to the reference mounting position. With the magnetic field sensor fixture 50 attached, for example, a magnetic marker is applied to the position of the nose or ear, magnetic field data of the magnetic marker is acquired by the sensor system, and this magnetic field data is used as reference magnetic field data for calibration. Record it.
And, for each case at the time of hospital and home treatment, the magnetic marker is applied to the position of the nose and ear, as in the case of acquiring the reference magnetic field data, and the magnetic field data of the magnetic marker is acquired by the sensor system, The mounting position of the magnetic field sensor fixture 50 in each case may be adjusted so that the acquired magnetic field data matches the reference magnetic field data as much as possible.
In addition, instead of separately setting the reference magnetic field data for calibration in this way, the magnetic field data of the magnet marker acquired at the hospital is recorded, and the magnetic field data of the magnetic marker acquired at home treatment is stored in the hospital. Calibration can also be performed by adjusting the mounting position of the magnetic field sensor fixture 50 at home treatment so that it matches the magnetic field data acquired in step 1 as much as possible.

For the method using the above data set, the time required for guidance to the final guidance position and posture corresponding to the optimal stimulation position and posture and the movement trajectory by the operation, the final guidance position and posture for the optimum stimulation position and posture Experiments were conducted to verify the effects of errors and the number of data sets.
The error of the final guidance position and posture with respect to the optimal stimulation position and posture was measured using POLARIS. As errors, the center position error, roll angle error, pitch angle error and yaw angle error of the stimulation coil were measured.

In this experiment, the mannequin head Hm was used instead of the patient's head, and the operator of the coil holder 12 was the subject. A plurality of subjects (for example, 3 people), all of whom were non-medical workers, conducted a guidance operation training for a predetermined time (for example, about several minutes) in order to get used to Open GL before starting the experiment. In addition, the non-medical worker (one of the inventors) performed the designation of the optimal stimulation position and posture and the collection of the data set in this experiment. In the system used in this experiment, the data sampling rate was, for example, 4 Hz.
In addition, the experiment was carried out with two patterns when the number of data sets was 500 and 1000. These data sets are obtained by focusing on the vicinity of the optimal stimulation position. The test subject performed 6 guidance operations, 3 times for each pattern.

As a result of the above experiment, the following items were confirmed.
B) Regardless of the subject, any error was greatly reduced when the number of data sets was 1000 compared to 500.
: In this method, even if the magnetic field generated by the permanent magnet does not completely match the magnetic field of the data set, the data set having the closest magnetic field is extracted and its position data is displayed.
That is, even if it is recognized that the guidance operation has been completed on the system, an error may occur from the data set designated as the optimum stimulation position. Therefore, when the number of data sets is small, it is considered that the error between the actual position of the stimulation coil and the position of the stimulation coil displayed on the Open GL increases on average.

B) Also, when the number of data sets was 1000, in most cases, it was within the following required specification error. It is assumed that the required specification error can be satisfied in all cases by further increasing the number of data sets.
The required specifications are preferably calculated based on clinical research. In this embodiment, as in the first embodiment, the coil center position and posture errors are, for example, as follows: Set to be within range.
-Coil center position error: within 5 [mm]-Coil posture error: within 5 [deg] for each coil axis

C) When the number of data sets is 1000, the time required for guidance is generally longer than when 500 data sets are used.
: In this experiment, when the same data set as the data set designated as the optimal stimulus position and posture is displayed on the Open GL, the termination condition of the guidance operation is set. Therefore, even if it is within the range of the required specification error, if the data sets do not match, the guidance operation will not end. If the number of data sets is large, the probability of being recognized as a data set that is not the data set of the optimal stimulus position and posture is increased accordingly, so that the guidance takes time.
However, in the case of this experiment, it takes only 55 seconds even in the longest example, and it can be said that this technique is sufficiently practical. Further, the time required for guidance can be expected to be shortened by the familiarity of the subject (or patient). Further, if it is adopted that the guide operation end condition is within the above-mentioned required specification error range, for example, further shortening can be expected.

As described above, reference is made to the comparison between the data on the magnetic field intensity and direction detected by the magnetic field sensor and the recording information (data set) of the recording unit 123 prior to or during the magnetic stimulation by the stimulation coil 11. Therefore, the position of the stimulation coil 11 is calculated and displayed on the user interface unit (display screen) 128, so that the magnetic stimulation apparatus can be reduced in size and size without using a complicated algorithm. This contributes to cost reduction and ease of operation. As a result, it is possible to allow the patient M to perform transcranial magnetic stimulation therapy on a daily basis at home or in a nearby doctor's office.

In addition, as described above, the effectiveness of the method using the data set was verified. However, when this method is adopted, it takes time and effort to collect the data set compared to the method using the inverse analysis of the magnetic field. Will do. For example, in the case of the experiment, since the data sampling rate is 4 Hz, for example, it takes about 5 minutes to collect 1000 data sets.
When actually applying a method using a data set to a medical site, a doctor needs to collect the data set for many points in the vicinity after identifying the optimal stimulus position of the patient's head. The set collection work is a burden on the doctor. Therefore, it is desirable to introduce a method for reducing the burden of data set collection work such as shortening the time required for data set collection.

-In order to shorten the time required for data set collection, it is advantageous if guidance with a smaller number of data sets can be made possible. One technique that makes this possible is to interpolate data sets by applying an interpolation method. As the interpolation method, for example, various known methods such as a method using multiple regression analysis and a method using weighted average can be applied.

A technique applying multiple regression analysis will be described as an example of such a data set interpolation technique.
For example, interpolation may be performed using the number of data sets of [number of sensors × 3 (axis) −2] (10 in the case of 4 sensors) out of n data sets. In this case, as the number of sensors increases, the reliability of interpolation increases.

Let A be magnetic field data acquired by the magnetic field sensor in real time (that is, magnetic field data at the position where the patient placed the coil). A is compared with magnetic field data B _i (i = 1, 2,..., N) of N data sets, and an f value expressed by Equation 22 is obtained for each data set (f ₁ to f _N ). .

However, A is a column vector of 12 rows and 1 column in which values in three directions of x, y, and z are arranged in order for the four magnetic field sensors a, b, c, and d, respectively, and B _i is Table 6 _Is a 12 × 1 column vector in which four three-dimensional vectors B _ai , B _bi , B _ci , B _di corresponding to the data set number i are vertically arranged.

Next, 10 sets having a small f value are extracted from the data set. A combination matrix of 12 rows and 10 columns in which the 10 sets of magnetic field data are arranged horizontally is represented by D. When the regression coefficient vector of 10 rows and 1 column is θ and the error vector of 12 rows and 1 column is ε, the magnetic field data A is expressed by Equation 23.

From this equation 23, the least square estimation amount Θ is obtained. Using this theta, the detection value of the magnetic field sensor is a position P _A and orientation R _A of the coil when the A, interpolated by the following Equation 24 and Equation 25.

However, P _i in Equation 24 and R _i in Equation 25 are coil position and orientation data corresponding to 10 sets of magnetic field data extracted from the data set.

Using the same method and conditions as in the previous experiment, the optimal stimulus position and posture were determined when the above-mentioned interpolation method was applied when the actual number of data sets was 500 and when the actual number of data sets was 1000. When an experiment was performed to compare the time required to induce the coil and the error after induction, both obtained substantially the same results, confirming that the above interpolation method was effective.

It is also conceivable to enable guidance with a smaller number of data sets by contriving the collection of data sets centered on the optimal stimulus position.
When using the data set to guide the coil to the optimal stimulation position and posture, there are many data sets near the optimal stimulation position in order to match the coil position and posture to the optimal stimulation position and posture as much as possible. Is desired. On the other hand, in a region greatly deviating from the optimal stimulation position, it is not necessary to collect as many data sets as in the vicinity of the optimal stimulation position.

Depending on the target disease of transcranial magnetic stimulation therapy, it may be clear which part of the brain is desirable to stimulate. For example, in the case of neuropathic pain, it is desirable to stimulate the “primary motor area” (see, for example, Patent Document 1: Paragraphs [0026], [0027], etc.). Further, the optimal stimulation site is generally limited (for example, in the case of neuropathic pain, the diameter does not exceed 20 mm, preferably 10 mm or less; see paragraph [0028] of Patent Document 1).

Therefore, from the viewpoint of data set collection, data in the vicinity of the optimal stimulation site is mainly collected at high density, whereas data in a region far from the optimal stimulation site is collected at relatively low density. It is efficient. That is, it is possible to reduce the number of collected data sets while maintaining the required time and accuracy when guiding the coil to the optimal stimulation position and posture.
The optimal stimulation site varies depending on the neurological disease that is assumed as a treatment target. In this case, it is important to collect data at a high density in a specific area corresponding to each neurological disease, while collecting data at a relatively low density in an area greatly deviating from this specific area. The same effect as the case can be produced.

Although it does not reduce the number of data sets, as one method for improving the efficiency of collecting data sets, the position of the patient's head can be identified by dividing it into a mesh and displaying the coordinates. It is also conceivable to make this easier.

Although it is conceivable to improve the efficiency of data set collection by applying the method described above, the work is still a burden on the doctor as long as the doctor collects the data set.
Therefore, the burden on the doctor can be reduced by performing the data set collection work separately for a doctor and a non-doctor.

More specifically, the doctor can identify the optimal stimulation position only, but the other data set collection work itself can be performed by a non-doctor. In other words, the optimal stimulation position needs to be identified by a doctor for a patient to be treated using a three-dimensional position measurement device (for example, an optical tracking system) at the hospital. On the other hand, data set collection work other than specifying the optimal stimulus position and posture is not a hospital, is not a 3D position measurement device installed in the hospital, and is not a doctor, even if there are no patients. But you can do it.

Therefore, the collection of the data set may be performed separately from the identification of the optimal stimulation position and posture by the doctor in the hospital. For example, it is preferable to carry out as part of the pre-shipment inspection of the manufacturer (manufacturer) of the magnetic stimulation apparatus.
That is, the data set is acquired in the coordinate system A on the manufacturer side using the manufacturer's three-dimensional position measuring apparatus. On the other hand, the optimal stimulus position and posture are acquired in another coordinate system B by using a three-dimensional position measuring device of the hospital at the hospital. Registration of the relationship between the coordinate system A and the coordinate system B can be easily performed by a known method as will be described later. Then, by converting the optimal stimulus position and orientation acquired in the coordinate system B into the coordinate system A, navigation using the data set acquired on the manufacturer side becomes possible. As a result, the doctor's burden associated with data collection is dramatically reduced.
In this case, the recording unit of the data set analysis unit has a function of recording information (preferably position and orientation) of at least the position of the magnetic field generation unit as position (and orientation) information of a plurality of different coordinate systems, It is preferable to have a coordinate conversion function that enables position reference (and posture) information in a plurality of different coordinate systems to be matched with each other and referenced for comparison.

A specific example of registration between the coordinate system A by the manufacturer's three-dimensional position measurement device and the coordinate system B by the hospital's three-dimensional position measurement device will be described. Also in this case, the method using the above-described

equations

1 and 2 can be applied. That is,
(1) First, four feature points that are not on the same plane that can be stably measured are determined on the magnetic field sensor fixture 50.
(2) Next, using the manufacturer's three-dimensional position measurement device, the position coordinates of the four feature points as seen in the coordinate system A:
(X ₁ , y ₁ , z ₁ ), (x ₂ , y ₂ , z ₂ ), (x ₃ , y ₃ , z ₃ ), (x ₄ , y ₄ , z ₄ )
To get.
(3) Further, the position coordinates of the same four feature points as seen in the coordinate system B using a hospital three-dimensional position measuring device:
(X ₁ , Y ₁ , Z ₁ ), (X ₂ , Y ₂ , Z ₂ ), (X ₃ , Y ₃ , Z ₃ ), (X ₄ , Y ₄ , Z ₄ )
To get.
(4) The coordinate transformation matrix T is calculated by the above equation 1.
(5) By using this coordinate transformation matrix T, the position of an arbitrary feature point (and hence the optimum stimulation position) acquired by the coordinate system B (hospital three-dimensional position measuring device) as shown in the equation 2 above. The coordinates (X, Y, Z) can be converted into position coordinates (x, y, z) as seen in the coordinate system A by the manufacturer's three-dimensional position measuring device, and navigation using the data set becomes possible.

In addition, when a data set (including values calculated by interpolation) is prepared at a sufficiently high density on the manufacturer side, this coordinate conversion process (that is, registration) itself can be made unnecessary. In other words, if the doctor determines the optimal stimulation position and posture at the hospital, referring to the data set prepared by the manufacturer, the three-dimensional position and posture corresponding to the magnetic field at that time (not the measured value at the hospital, The value called from the data set or the value calculated by interpolation may be used as the optimum stimulation position and posture as it is. In this case, the coordinate conversion process described above is performed using the data set.

The above procedure will be described more specifically.
<Manufacturer>: For example, during pre-shipment inspection a) A magnetic field sensor fixture (for example, glasses with a magnetic field sensor) is attached to a head model based on a standard adult head structure.
B) Using a three-dimensional position measurement device (coordinate system A) installed at the manufacturer, a magnetic field generated by a permanent magnet attached to a coil holder together with a stimulation coil (measured by a magnetic field sensor attached to glasses), and a stimulation coil A combination (data set) of the three-dimensional position and orientation in the coordinate system A is collected while moving the coil holder. At this time, as many data sets as possible are collected in consideration of the difference between the head model and the actual patient's head, and the occurrence of a shift in the wearing position of the glasses.

<At the hospital>: During initial treatment a) Have the patient wear a magnetic field sensor fixture (for example, glasses with a magnetic field sensor).
B) Magnetic field generated by a permanent magnet attached to a coil holder together with a stimulation coil (measured by a magnetic field sensor attached to glasses) using a three-dimensional position measuring device (coordinate system B) installed in a hospital, and a stimulation coil While moving the coil holder, the combination of the three-dimensional position and orientation in the coordinate system A (estimated from the data set obtained by the manufacturer) and the three-dimensional position and orientation in the coordinate system B of the stimulation coil, Collect several points (at least 4 points).
C) From the correspondence between the coordinate system A and the coordinate system B at these points, the coordinate transformation matrix of the coordinate system A and the coordinate system B is obtained by applying the above-described registration.
D) Next, the doctor specifies the optimal stimulus position and posture in the coordinate system B.
E) The optimal stimulus position and posture are converted into the coordinate system A and recorded using the coordinate conversion matrix obtained in the item c).

<At home>
Since the data set (coordinate system A) acquired by the manufacturer is converted into the same coordinates (coordinate system A) and the data set of the optimal stimulus position and posture acquired at the hospital is additionally recorded, the patient can Navigation operations can be performed.
As described above, the burden on the doctor can be minimized without changing the framework of the technique using the data set.

Here, a method of applying the above-described “method for effectively collecting a data set centering on the optimal stimulus position” at the time of data set collection on the manufacturer side will be described.
For example, as described above, depending on the target disease of transcranial magnetic stimulation therapy, it may be clear which part of the brain is desired to be stimulated. Therefore, when collecting the data set on the manufacturer side, roughly grasp the area where the optimal stimulation site will be located on the head model, and collect data at a higher density while focusing on that area. By collecting data at a relatively low density in an area greatly deviating from the area, efficient data set collection can be performed.

In addition, after acquiring information about the patient's head and optimal stimulation position at the hospital, the information can be sent to the manufacturer, and based on that information, the manufacturer can collect a data set tailored to the patient. Conceivable. In this case, the manufacturer collects data near the optimal stimulation position with a higher density based on the position information of the optimal stimulation position, while collecting data at a relatively low density in a region greatly deviating from the optimal stimulation position. You just have to. In this case, efficient data set collection can be performed by obtaining position information regarding the optimal stimulus position from the hospital (even roughly).

In the above description, the acquisition of the data set including the data set of the optimal stimulus position and orientation is performed exclusively using the three-dimensional position measuring device such as the optical tracking device. Instead, the first embodiment is used. It is also possible to acquire a data set by using the magnetic field inverse analysis method described in.

It is also possible to combine the method using the inverse analysis method of the magnetic field and the method using the data set so as to guide the stimulation coil to the optimal stimulation position and posture more efficiently.
In this case, for example, in a region where a data set has not been acquired (that is, a region far away from the optimal stimulation position to some extent), a stimulation coil is induced by a method using a magnetic field inverse analysis method, and data In areas where sets are acquired (that is, areas that are relatively close to the optimal stimulation position), it is possible to induce stimulation coils using a method that uses data sets, so that the number of data sets is relatively small Efficient and smooth induction of the stimulation coil is possible.

As described above, when the coil is guided to the optimal stimulation position and posture using the data set, it is desirable that the number of data sets is large in order to match the coil position and posture to the optimal stimulation position and posture as much as possible. In particular, the number of data sets in the vicinity of the optimal stimulation position affects the accuracy of the coil guiding position and posture. Therefore, it is also conceivable to apply an inverse analysis method in the final process of coil induction, instead of performing coil induction only with the data set to the end. In this case, it is possible to reduce the number of data sets near the optimal stimulus position to some extent.
In addition, in the initial stage of the coil induction process, the inverse analysis method is applied in an area where the coil position is a certain distance or more from the optimal stimulation position and the data set is not acquired. When the acquired area is reached, rough alignment is performed using the data set method, and in the final coil induction process, it is possible to apply the inverse analysis method to perform final alignment. .

In the above embodiment, basically, the stimulation coil is operated and relatively displaced with respect to the fixed patient's head. Instead, the stimulation coil is fixed and fixed. The patient's head can be relatively displaced with respect to the prepared stimulation coil. Alternatively, the present invention can be applied effectively even when both the patient's head and the stimulation coil are operated and relatively displaced.

As an example of the embodiment, the positioning procedure to the optimum stimulation position at the time of treatment at home is exemplified as follows.
First, as in the previous embodiments, a magnetic field sensor as a magnetic field detection means is fixed to the patient's head using fixing means such as glasses. On the other hand, the stimulation coil is fixed by the holder fixture so as to face the rough stimulation position of the head (for example, a region corresponding to the primary motor area).

After completion of the above setting, positioning may be basically performed according to a process corresponding to the process of the flowchart shown in FIG. If it demonstrates using the flowchart of FIG. 5, in this embodiment, a patient will move his head so that it may suit an optimal position in step # 22. In step # 23, the deviation (deviation) from the optimal position and posture of the stimulation coil is detected as in the previous embodiments. In step # 24, the user interface unit notifies the patient how to move his / her head. That is, the patient's head movement is navigated so that the stimulation coil has the optimum position and posture. Through this process, the treatment at the optimal stimulation position is possible as in the previous embodiments.

In step # 22, the patient may move his / her head and also perform the displacement operation for the stimulation coil. In this case, the user interface unit functions as a notification unit for guiding the displacement operation of the stimulation coil and a notification unit for guiding the movement of the patient's head, so that the optimal stimulation of the stimulation coil is performed. Effective navigation to the position and posture can be performed.

The above embodiments are all for transcranial magnetic stimulation therapy in which nerve stimulation is applied to intracerebral nerves by a stimulation coil placed on the surface of the patient's scalp to alleviate neuropathic pain. However, the present invention is not limited to such a case, and can be effectively applied to other magnetic stimulation applications.

Thus, it goes without saying that the present invention is not limited to the above-described embodiments, and various changes and design improvements can be made without departing from the scope of the invention.

The present invention relates to a magnetic stimulation apparatus for applying magnetic stimulation to a specific part of a subject, for example, an apparatus used for transcranial magnetic stimulation therapy in which magnetic stimulation is applied to, for example, brain nerves by a stimulation coil arranged on the surface of a patient's scalp. As such, it can be used effectively.

DESCRIPTION OF SYMBOLS 10 Transcranial magnetic stimulation apparatus 11 Stimulation coil 12 Coil holder 13 Magnetic field sensor 14 Glasses 15 Cable 16 Magnetic stimulation control apparatus 20 Magnetic field analysis unit 22,122 Signal analysis part 23 Storage part 24,124 Comparison part 25,125 User information output part 28,128 User interface unit 41 Permanent magnet 50 Magnetic field sensor fixture 51 Magnetic field sensor 120 Data set analysis unit 123 Recording unit M Patient

Claims

A magnetic stimulation device for applying magnetic stimulation to a specific part of a subject,
Magnetic field generating means including at least dynamic magnetic field generating means for generating a dynamic magnetic field for applying the magnetic stimulation;
An operation means operated so as to be able to displace a relative position of the magnetic field generating means with respect to a specific part of the subject;
A plurality of magnetic field detecting means for detecting the magnetic field generated by the magnetic field generating means;
Teaching for teaching the displacement operation to be performed using the operation means based on the detection result of the magnetic field generated from the magnetic field generation means prior to or during the magnetic stimulation by the magnetic field detection means. An informing means for informing the information;
A magnetic stimulation apparatus comprising:
A magnetic stimulation device for applying magnetic stimulation to a specific part of a subject,
Magnetic field generating means including at least dynamic magnetic field generating means for generating a dynamic magnetic field for applying the magnetic stimulation;
Holding means for holding the magnetic field generating means in the vicinity of a specific part of the subject;
A plurality of magnetic field detecting means for detecting the magnetic field generated by the magnetic field generating means;
Prior to or during the magnetic stimulation, based on the result of detection of the magnetic field generated by the magnetic field generation means by the magnetic field detection means, the subject should perform the body movement for the magnetic stimulation to the specific site. Informing means for informing teaching information for teaching
A magnetic stimulation apparatus comprising:
The magnetic stimulation apparatus according to claim 1 or 2, further comprising a fixing means for fixing the magnetic field detection means at a predetermined relative position with respect to the specific part of the subject.
4. The magnetic stimulation apparatus according to claim 1, wherein the magnetic field generating means is attached to the operating means.
The notification means calculates the position of the magnetic field generation means as the position of the magnetic field source obtained by an inverse analysis method using each information on the magnetic field strength and direction detected by the plurality of magnetic field detection means, and the teaching information The magnetic stimulation apparatus according to any one of claims 1 to 4, wherein the magnetic stimulation apparatus generates and informs.
A record that is recorded in advance at a plurality of the at least positions by pairing information on the position of the magnetic field generating means and information on the magnetic field intensity and direction detected by the magnetic field detecting means for the magnetic field generated at the position. Further comprising means,
The informing means generates the magnetic field based on a comparison reference between each piece of information on the magnetic field strength and direction detected by each magnetic field detecting means prior to or during the magnetic stimulation and the recording information of the recording means. 5. The magnetic stimulation apparatus according to claim 1, wherein a position of means is calculated to generate and notify the teaching information.
Each information regarding the magnetic field intensity and direction detected by each magnetic field detection means in a state where the magnetic field generation means is located within a position for applying a magnetic stimulus to a specific part of the subject or within an allowable vicinity thereof. A plurality of pre-recorded target information recording means, wherein the notification means includes information relating to the magnetic field strength and direction detected by the magnetic field detection means prior to or during the magnetic stimulation, and the target information recording means. 5. The magnetic stimulation apparatus according to claim 1, wherein the teaching information is generated and notified based on a comparison reference result with the recorded information.
The recording means can record the position information of the magnetic field generating means as position information in a plurality of different coordinate systems, and match the position information in the different coordinate systems with each other and refer to the comparison The magnetic stimulation apparatus according to claim 6, further comprising coordinate conversion means for enabling
9. The magnetic stimulation apparatus according to claim 1, wherein the magnetic field generating means generates a dynamic magnetic field and a static magnetic field.
9. The magnetic stimulation apparatus according to claim 1, wherein the magnetic field generating means generates only a dynamic magnetic field.
10. The magnetic stimulation apparatus according to claim 9, wherein the magnetic field detection means detects a dynamic magnetic field and a static magnetic field generated by the magnetic field generation means.
12. The magnetic stimulation apparatus according to claim 11, wherein the magnetic field detection means detects a static magnetic field by the magnetic field generation means in a state where generation of a dynamic magnetic field by the magnetic field generation means is stopped.
The magnetic stimulation apparatus according to claim 9 or 10, wherein the magnetic field detection means detects only a dynamic magnetic field generated by the magnetic field generation means.
14. The magnetic stimulation apparatus according to claim 1, wherein the notifying unit notifies at least one of visual information and auditory information.
The informing means is an informing means for informing the teaching information by auditory information, and at least one of a volume, a scale, and a timbre according to a displacement amount to be performed by the operation means or a movement amount to be moved by the subject. The magnetic stimulation device according to claim 14, wherein the magnetic stimulation device is changed.
The notifying means is notifying means for notifying the teaching information by visual information, and the teaching color is changed according to a displacement amount to be performed by the operating means or a movement amount to be moved by the subject. The magnetic stimulation apparatus according to claim 14.
17. The magnetic stimulation apparatus according to claim 3, wherein the fixing means is selected from glasses, earphones, headphones, and a headband.
The magnetic stimulation device according to any one of claims 1 to 17, wherein the magnetic stimulation device applies magnetic stimulation to at least a specific part of a brain of a subject for transcranial magnetic stimulation treatment.